From 0f2d404aa1414a89c49bdcfde342936381e16196 Mon Sep 17 00:00:00 2001
From: artur-sannikov <artur-sannikov@users.noreply.github.com>
Date: Thu, 24 Oct 2024 13:31:11 +0000
Subject: [PATCH] =?UTF-8?q?Deploying=20to=20gh-pages=20from=20@=20EBI-Meta?=
 =?UTF-8?q?genomics/HoloFoodR@0da8d07e23e0e15236d065eec59c3cd4971fa1c5=20?=
 =?UTF-8?q?=F0=9F=9A=80?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 404.html                                      |   2 +-
 LICENSE-text.html                             |   2 +-
 articles/HoloFoodR.html                       |   4 +-
 articles/case_study.html                      | 222 +++++++++---------
 .../figure-html/abundance_plot-1.png          | Bin 70579 -> 74234 bytes
 articles/index.html                           |   2 +-
 authors.html                                  |   6 +-
 index.html                                    |   2 +-
 pkgdown.yml                                   |   2 +-
 reference/HoloFoodR.html                      |   2 +-
 reference/addMGnify.html                      |   2 +-
 reference/doQuery.html                        |   2 +-
 reference/getData.html                        |   2 +-
 reference/getMetaboLights.html                |   2 +-
 reference/getResult.html                      |   2 +-
 reference/index.html                          |   2 +-
 search.json                                   |   2 +-
 17 files changed, 130 insertions(+), 128 deletions(-)

diff --git a/404.html b/404.html
index 00c568c..9cde712 100644
--- a/404.html
+++ b/404.html
@@ -20,7 +20,7 @@
 
     <a class="navbar-brand me-2" href="index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/LICENSE-text.html b/LICENSE-text.html
index e762e4f..7d077da 100644
--- a/LICENSE-text.html
+++ b/LICENSE-text.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/articles/HoloFoodR.html b/articles/HoloFoodR.html
index d244995..9c816cf 100644
--- a/articles/HoloFoodR.html
+++ b/articles/HoloFoodR.html
@@ -20,7 +20,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -418,7 +418,7 @@ <h2 id="session-info">Session info<a class="anchor" aria-label="anchor" href="#s
 <span><span class="co">#&gt; [8] base     </span></span>
 <span><span class="co">#&gt; </span></span>
 <span><span class="co">#&gt; other attached packages:</span></span>
-<span><span class="co">#&gt;  [1] HoloFoodR_0.99.10               TreeSummarizedExperiment_2.13.0</span></span>
+<span><span class="co">#&gt;  [1] HoloFoodR_0.99.11               TreeSummarizedExperiment_2.13.0</span></span>
 <span><span class="co">#&gt;  [3] Biostrings_2.73.2               XVector_0.45.0                 </span></span>
 <span><span class="co">#&gt;  [5] SingleCellExperiment_1.27.2     MultiAssayExperiment_1.31.5    </span></span>
 <span><span class="co">#&gt;  [7] SummarizedExperiment_1.35.4     Biobase_2.65.1                 </span></span>
diff --git a/articles/case_study.html b/articles/case_study.html
index 72821c6..7c9f9ca 100644
--- a/articles/case_study.html
+++ b/articles/case_study.html
@@ -20,7 +20,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -99,7 +99,7 @@ <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#i
 <div class="sourceCode" id="cb2"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co"># List of packages that we need</span></span>
 <span><span class="va">packages</span> <span class="op">&lt;-</span> <span class="fu"><a href="https://rdrr.io/r/base/c.html" class="external-link">c</a></span><span class="op">(</span></span>
-<span>    <span class="st">"dplyr"</span>, <span class="st">"DT"</span>, <span class="st">"ggsignif"</span>, <span class="st">"HoloFoodR"</span>, <span class="st">"MGnifyR"</span>, <span class="st">"mia"</span>, <span class="st">"miaViz"</span>, <span class="st">"MOFA2"</span>,</span>
+<span>    <span class="st">"dplyr"</span>, <span class="st">"DT"</span>, <span class="st">"ggpubr"</span>,  <span class="st">"ggsignif"</span>, <span class="st">"HoloFoodR"</span>, <span class="st">"MGnifyR"</span>, <span class="st">"mia"</span>, <span class="st">"miaViz"</span>, <span class="st">"MOFA2"</span>,</span>
 <span>    <span class="st">"patchwork"</span>, <span class="st">"reticulate"</span>, <span class="st">"scater"</span></span>
 <span><span class="op">)</span></span>
 <span></span>
@@ -116,6 +116,8 @@ <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#i
 <span><span class="co">#&gt; </span></span>
 <span><span class="co">#&gt;     intersect, setdiff, setequal, union</span></span>
 <span><span class="co">#&gt; Loading required package: DT</span></span>
+<span><span class="co">#&gt; Loading required package: ggpubr</span></span>
+<span><span class="co">#&gt; Loading required package: ggplot2</span></span>
 <span><span class="co">#&gt; Loading required package: ggsignif</span></span>
 <span><span class="co">#&gt; Loading required package: HoloFoodR</span></span>
 <span><span class="co">#&gt; Loading required package: MultiAssayExperiment</span></span>
@@ -223,7 +225,6 @@ <h2 id="introduction">Introduction<a class="anchor" aria-label="anchor" href="#i
 <span><span class="co">#&gt; </span></span>
 <span><span class="co">#&gt;     full_join, inner_join, left_join, right_join</span></span>
 <span><span class="co">#&gt; Loading required package: miaViz</span></span>
-<span><span class="co">#&gt; Loading required package: ggplot2</span></span>
 <span><span class="co">#&gt; Loading required package: ggraph</span></span>
 <span><span class="co">#&gt; </span></span>
 <span><span class="co">#&gt; Attaching package: 'miaViz'</span></span>
@@ -693,11 +694,9 @@ <h3 id="microbial-composition">Microbial composition<a class="anchor" aria-label
 <span>    facet.cols <span class="op">=</span> <span class="cn">TRUE</span>, scales <span class="op">=</span> <span class="st">"free_x"</span></span>
 <span>    <span class="op">)</span></span>
 <span><span class="va">p</span></span>
-<span><span class="co">#&gt; $abundance</span></span></code></pre></div>
+<span><span class="co">#&gt; Warning: Removed 200 rows containing missing values or values outside the scale range</span></span>
+<span><span class="co">#&gt; (`geom_bar()`).</span></span></code></pre></div>
 <p><img src="case_study_files/figure-html/abundance_plot-1.png" width="700"></p>
-<pre><code><span><span class="co">#&gt; </span></span>
-<span><span class="co">#&gt; $treatment_group</span></span></code></pre>
-<p><img src="case_study_files/figure-html/abundance_plot-2.png" width="700"></p>
 <p>Salmon gut seems to be dominated by either genus <em>Mycoplasma</em>
 or <em>Photobacterium</em>.</p>
 </div>
@@ -706,7 +705,7 @@ <h3 id="is-alpha-diversity-associated-with-treatment-or-mass-of-salmon">Is alpha
 </h3>
 <p>Next, let’s calculate the Shannon alpha diversity index and test
 whether diversity is associated with both size and treatment.</p>
-<div class="sourceCode" id="cb28"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb27"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co"># Calculate alpha diversity</span></span>
 <span><span class="va">mae</span><span class="op">[[</span><span class="fl">2</span><span class="op">]</span><span class="op">]</span> <span class="op">&lt;-</span> <span class="fu">addAlpha</span><span class="op">(</span><span class="va">mae</span><span class="op">[[</span><span class="fl">2</span><span class="op">]</span><span class="op">]</span><span class="op">)</span></span>
 <span><span class="co">#&gt; Warning: 'faith' index can be calculated only for TreeSE with rowTree(x)</span></span>
@@ -732,7 +731,7 @@ <h3 id="is-alpha-diversity-associated-with-treatment-or-mass-of-salmon">Is alpha
 <p>Let’s analyze whether we can find similar effect with beta diversity.
 Here we perform Principal Coordinate Analysis (PCoA) with Bray-Curtis
 dissimilarity.</p>
-<div class="sourceCode" id="cb29"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb28"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="co"># Run PCoA</span></span>
 <span><span class="va">mae</span><span class="op">[[</span><span class="fl">2</span><span class="op">]</span><span class="op">]</span> <span class="op">&lt;-</span> <span class="fu">runMDS</span><span class="op">(</span></span>
 <span>    <span class="va">mae</span><span class="op">[[</span><span class="fl">2</span><span class="op">]</span><span class="op">]</span>,</span>
@@ -763,7 +762,7 @@ <h2 id="multi-omics-integration">Multi-omics integration<a class="anchor" aria-l
 metagenomics and fatty acids exhibit shared variability, ultimately
 assessing whether the microbial community is associated with fatty acid
 composition.</p>
-<div class="sourceCode" id="cb30"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb29"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">mae_temp</span> <span class="op">&lt;-</span> <span class="va">mae</span></span>
 <span><span class="va">mae_temp</span><span class="op">[[</span><span class="fl">2</span><span class="op">]</span><span class="op">]</span> <span class="op">&lt;-</span> <span class="fu">altExp</span><span class="op">(</span><span class="va">mae_temp</span><span class="op">[[</span><span class="fl">2</span><span class="op">]</span><span class="op">]</span>, <span class="st">"prev_genus"</span><span class="op">)</span></span>
 <span></span>
@@ -801,11 +800,11 @@ <h2 id="multi-omics-integration">Multi-omics integration<a class="anchor" aria-l
 <span></span>
 <span><span class="co"># Train model</span></span>
 <span><span class="va">model</span> <span class="op">&lt;-</span> <span class="fu">run_mofa</span><span class="op">(</span><span class="va">model</span>, use_basilisk <span class="op">=</span> <span class="cn">TRUE</span><span class="op">)</span></span>
-<span><span class="co">#&gt; Warning in run_mofa(model, use_basilisk = TRUE): No output filename provided. Using /tmp/RtmpGcJK0u/mofa_20241024-125203.hdf5 to store the trained model.</span></span>
+<span><span class="co">#&gt; Warning in run_mofa(model, use_basilisk = TRUE): No output filename provided. Using /tmp/RtmpMdYbfu/mofa_20241024-132828.hdf5 to store the trained model.</span></span>
 <span><span class="co">#&gt; Connecting to the mofapy2 package using basilisk. </span></span>
 <span><span class="co">#&gt;     Set 'use_basilisk' to FALSE if you prefer to manually set the python binary using 'reticulate'.</span></span></code></pre></div>
 <p>Next, we will plot the variances explained by each factor.</p>
-<div class="sourceCode" id="cb31"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb30"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">p</span> <span class="op">&lt;-</span> <span class="fu">plot_variance_explained</span><span class="op">(</span><span class="va">model</span><span class="op">)</span> <span class="op">+</span></span>
 <span>    <span class="fu">labs</span><span class="op">(</span>title <span class="op">=</span> <span class="st">"Captured variance of fatty acid and metagenomic\n"</span>,</span>
 <span>        <span class="st">"datasets across 5 factors"</span><span class="op">)</span> <span class="op">+</span></span>
@@ -819,7 +818,7 @@ <h2 id="multi-omics-integration">Multi-omics integration<a class="anchor" aria-l
 be captured by the second factor.</p>
 <p>Before exploring the shared variability, we first examine which
 taxa’s variability is captured by Factor 1.</p>
-<div class="sourceCode" id="cb32"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb31"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">p1</span> <span class="op">&lt;-</span> <span class="fu">plot_top_weights</span><span class="op">(</span><span class="va">model</span>, view <span class="op">=</span> <span class="fl">1</span>, factors <span class="op">=</span> <span class="fl">1</span>, nfeatures <span class="op">=</span> <span class="fl">25</span><span class="op">)</span> <span class="op">+</span></span>
 <span>  <span class="fu">labs</span><span class="op">(</span>title <span class="op">=</span> <span class="st">"Fatty acids"</span><span class="op">)</span></span>
 <span><span class="va">p2</span> <span class="op">&lt;-</span> <span class="fu">plot_top_weights</span><span class="op">(</span><span class="va">model</span>, view <span class="op">=</span> <span class="fl">2</span>, factors <span class="op">=</span> <span class="fl">1</span>, nfeatures <span class="op">=</span> <span class="fl">25</span><span class="op">)</span> <span class="op">+</span></span>
@@ -829,7 +828,7 @@ <h2 id="multi-omics-integration">Multi-omics integration<a class="anchor" aria-l
 <p><img src="case_study_files/figure-html/plot_factor1-1.png" width="960"></p>
 <p>From the plot above, we can see that the first factor captures mostly
 the variability in <em>Mycoplasma</em>.</p>
-<div class="sourceCode" id="cb33"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb32"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="va">p1</span> <span class="op">&lt;-</span> <span class="fu">plot_top_weights</span><span class="op">(</span><span class="va">model</span>, view <span class="op">=</span> <span class="fl">1</span>, factors <span class="op">=</span> <span class="fl">2</span>, nfeatures <span class="op">=</span> <span class="fl">25</span><span class="op">)</span> <span class="op">+</span></span>
 <span>  <span class="fu">labs</span><span class="op">(</span>title <span class="op">=</span> <span class="st">"Fatty acids"</span><span class="op">)</span></span>
 <span><span class="va">p2</span> <span class="op">&lt;-</span> <span class="fu">plot_top_weights</span><span class="op">(</span><span class="va">model</span>, view <span class="op">=</span> <span class="fl">2</span>, factors <span class="op">=</span> <span class="fl">2</span>, nfeatures <span class="op">=</span> <span class="fl">25</span><span class="op">)</span> <span class="op">+</span></span>
@@ -853,7 +852,7 @@ <h2 id="conclusions">Conclusions<a class="anchor" aria-label="anchor" href="#con
 access to an extensive plethora of downstream tools, such
 <code>mia</code> and <code>MOFA2</code> that can be used to pre-process
 and visualize the multi-omics data.</p>
-<div class="sourceCode" id="cb34"><pre class="downlit sourceCode r">
+<div class="sourceCode" id="cb33"><pre class="downlit sourceCode r">
 <code class="sourceCode R"><span><span class="fu"><a href="https://rdrr.io/r/utils/sessionInfo.html" class="external-link">sessionInfo</a></span><span class="op">(</span><span class="op">)</span></span>
 <span><span class="co">#&gt; R version 4.4.1 (2024-06-14)</span></span>
 <span><span class="co">#&gt; Platform: x86_64-pc-linux-gnu</span></span>
@@ -881,103 +880,106 @@ <h2 id="conclusions">Conclusions<a class="anchor" aria-label="anchor" href="#con
 <span><span class="co">#&gt; other attached packages:</span></span>
 <span><span class="co">#&gt;  [1] scater_1.33.4                   scuttle_1.15.4                 </span></span>
 <span><span class="co">#&gt;  [3] reticulate_1.39.0               patchwork_1.3.0                </span></span>
-<span><span class="co">#&gt;  [5] MOFA2_1.15.0                    miaViz_1.13.14                 </span></span>
-<span><span class="co">#&gt;  [7] ggraph_2.2.1                    ggplot2_3.5.1                  </span></span>
-<span><span class="co">#&gt;  [9] mia_1.13.47                     MGnifyR_1.1.0                  </span></span>
-<span><span class="co">#&gt; [11] HoloFoodR_0.99.10               TreeSummarizedExperiment_2.13.0</span></span>
-<span><span class="co">#&gt; [13] Biostrings_2.73.2               XVector_0.45.0                 </span></span>
-<span><span class="co">#&gt; [15] SingleCellExperiment_1.27.2     MultiAssayExperiment_1.31.5    </span></span>
-<span><span class="co">#&gt; [17] SummarizedExperiment_1.35.4     Biobase_2.65.1                 </span></span>
-<span><span class="co">#&gt; [19] GenomicRanges_1.57.2            GenomeInfoDb_1.41.2            </span></span>
-<span><span class="co">#&gt; [21] IRanges_2.39.2                  S4Vectors_0.43.2               </span></span>
-<span><span class="co">#&gt; [23] BiocGenerics_0.51.3             MatrixGenerics_1.17.0          </span></span>
-<span><span class="co">#&gt; [25] matrixStats_1.4.1               ggsignif_0.6.4                 </span></span>
-<span><span class="co">#&gt; [27] DT_0.33                         dplyr_1.1.4                    </span></span>
-<span><span class="co">#&gt; [29] knitr_1.48                      BiocStyle_2.33.1               </span></span>
+<span><span class="co">#&gt;  [5] MOFA2_1.15.0                    miaViz_1.13.15                 </span></span>
+<span><span class="co">#&gt;  [7] ggraph_2.2.1                    mia_1.13.47                    </span></span>
+<span><span class="co">#&gt;  [9] MGnifyR_1.1.0                   HoloFoodR_0.99.11              </span></span>
+<span><span class="co">#&gt; [11] TreeSummarizedExperiment_2.13.0 Biostrings_2.73.2              </span></span>
+<span><span class="co">#&gt; [13] XVector_0.45.0                  SingleCellExperiment_1.27.2    </span></span>
+<span><span class="co">#&gt; [15] MultiAssayExperiment_1.31.5     SummarizedExperiment_1.35.4    </span></span>
+<span><span class="co">#&gt; [17] Biobase_2.65.1                  GenomicRanges_1.57.2           </span></span>
+<span><span class="co">#&gt; [19] GenomeInfoDb_1.41.2             IRanges_2.39.2                 </span></span>
+<span><span class="co">#&gt; [21] S4Vectors_0.43.2                BiocGenerics_0.51.3            </span></span>
+<span><span class="co">#&gt; [23] MatrixGenerics_1.17.0           matrixStats_1.4.1              </span></span>
+<span><span class="co">#&gt; [25] ggsignif_0.6.4                  ggpubr_0.6.0                   </span></span>
+<span><span class="co">#&gt; [27] ggplot2_3.5.1                   DT_0.33                        </span></span>
+<span><span class="co">#&gt; [29] dplyr_1.1.4                     knitr_1.48                     </span></span>
+<span><span class="co">#&gt; [31] BiocStyle_2.33.1               </span></span>
 <span><span class="co">#&gt; </span></span>
 <span><span class="co">#&gt; loaded via a namespace (and not attached):</span></span>
-<span><span class="co">#&gt;   [1] splines_4.4.1               filelock_1.0.3             </span></span>
-<span><span class="co">#&gt;   [3] ggplotify_0.1.2             urltools_1.7.3             </span></span>
-<span><span class="co">#&gt;   [5] tibble_3.2.1                triebeard_0.4.1            </span></span>
-<span><span class="co">#&gt;   [7] polyclip_1.10-7             basilisk.utils_1.17.3      </span></span>
-<span><span class="co">#&gt;   [9] rpart_4.1.23                DirichletMultinomial_1.47.2</span></span>
-<span><span class="co">#&gt;  [11] lifecycle_1.0.4             httr2_1.0.5                </span></span>
-<span><span class="co">#&gt;  [13] lattice_0.22-6              MASS_7.3-61                </span></span>
-<span><span class="co">#&gt;  [15] crosstalk_1.2.1             SnowballC_0.7.1            </span></span>
-<span><span class="co">#&gt;  [17] backports_1.5.0             magrittr_2.0.3             </span></span>
-<span><span class="co">#&gt;  [19] Hmisc_5.1-3                 sass_0.4.9                 </span></span>
-<span><span class="co">#&gt;  [21] rmarkdown_2.28              jquerylib_0.1.4            </span></span>
-<span><span class="co">#&gt;  [23] yaml_2.3.10                 RColorBrewer_1.1-3         </span></span>
-<span><span class="co">#&gt;  [25] cowplot_1.1.3               DBI_1.2.3                  </span></span>
-<span><span class="co">#&gt;  [27] minqa_1.2.8                 abind_1.4-8                </span></span>
-<span><span class="co">#&gt;  [29] zlibbioc_1.51.2             Rtsne_0.17                 </span></span>
-<span><span class="co">#&gt;  [31] purrr_1.0.2                 yulab.utils_0.1.7          </span></span>
-<span><span class="co">#&gt;  [33] nnet_7.3-19                 tweenr_2.0.3               </span></span>
-<span><span class="co">#&gt;  [35] rappdirs_0.3.3              sandwich_3.1-1             </span></span>
-<span><span class="co">#&gt;  [37] GenomeInfoDbData_1.2.13     ggrepel_0.9.6              </span></span>
-<span><span class="co">#&gt;  [39] tokenizers_0.3.0            irlba_2.3.5.1              </span></span>
-<span><span class="co">#&gt;  [41] tidytree_0.4.6              pheatmap_1.0.12            </span></span>
-<span><span class="co">#&gt;  [43] vegan_2.6-8                 rbiom_1.0.3                </span></span>
-<span><span class="co">#&gt;  [45] tidyjson_0.3.2              pkgdown_2.1.1              </span></span>
-<span><span class="co">#&gt;  [47] permute_0.9-7               DelayedMatrixStats_1.27.3  </span></span>
-<span><span class="co">#&gt;  [49] codetools_0.2-20            DelayedArray_0.31.14       </span></span>
-<span><span class="co">#&gt;  [51] ggforce_0.4.2               tidyselect_1.2.1           </span></span>
-<span><span class="co">#&gt;  [53] aplot_0.2.3                 UCSC.utils_1.1.0           </span></span>
-<span><span class="co">#&gt;  [55] farver_2.1.2                lme4_1.1-35.5              </span></span>
-<span><span class="co">#&gt;  [57] ScaledMatrix_1.13.0         viridis_0.6.5              </span></span>
-<span><span class="co">#&gt;  [59] base64enc_0.1-3             jsonlite_1.8.9             </span></span>
-<span><span class="co">#&gt;  [61] BiocNeighbors_1.99.3        decontam_1.25.0            </span></span>
-<span><span class="co">#&gt;  [63] tidygraph_1.3.1             Formula_1.2-5              </span></span>
-<span><span class="co">#&gt;  [65] systemfonts_1.1.0           ggnewscale_0.5.0           </span></span>
-<span><span class="co">#&gt;  [67] tools_4.4.1                 treeio_1.29.1              </span></span>
-<span><span class="co">#&gt;  [69] ragg_1.3.3                  Rcpp_1.0.13                </span></span>
-<span><span class="co">#&gt;  [71] glue_1.8.0                  BiocBaseUtils_1.7.3        </span></span>
-<span><span class="co">#&gt;  [73] gridExtra_2.3               SparseArray_1.5.45         </span></span>
-<span><span class="co">#&gt;  [75] xfun_0.48                   mgcv_1.9-1                 </span></span>
-<span><span class="co">#&gt;  [77] HDF5Array_1.33.8            withr_3.0.1                </span></span>
-<span><span class="co">#&gt;  [79] BiocManager_1.30.25         fastmap_1.2.0              </span></span>
-<span><span class="co">#&gt;  [81] basilisk_1.17.2             rhdf5filters_1.17.0        </span></span>
-<span><span class="co">#&gt;  [83] boot_1.3-31                 bluster_1.15.1             </span></span>
-<span><span class="co">#&gt;  [85] fansi_1.0.6                 digest_0.6.37              </span></span>
-<span><span class="co">#&gt;  [87] rsvd_1.0.5                  gridGraphics_0.5-1         </span></span>
-<span><span class="co">#&gt;  [89] R6_2.5.1                    textshaping_0.4.0          </span></span>
-<span><span class="co">#&gt;  [91] colorspace_2.1-1            lpSolve_5.6.21             </span></span>
-<span><span class="co">#&gt;  [93] UpSetR_1.4.0                utf8_1.2.4                 </span></span>
-<span><span class="co">#&gt;  [95] tidyr_1.3.1                 generics_0.1.3             </span></span>
-<span><span class="co">#&gt;  [97] data.table_1.16.2           DECIPHER_3.1.5             </span></span>
-<span><span class="co">#&gt;  [99] graphlayouts_1.2.0          httr_1.4.7                 </span></span>
-<span><span class="co">#&gt; [101] htmlwidgets_1.6.4           S4Arrays_1.5.11            </span></span>
-<span><span class="co">#&gt; [103] uwot_0.2.2                  pkgconfig_2.0.3            </span></span>
-<span><span class="co">#&gt; [105] gtable_0.3.5                janeaustenr_1.0.0          </span></span>
-<span><span class="co">#&gt; [107] htmltools_0.5.8.1           bookdown_0.41              </span></span>
-<span><span class="co">#&gt; [109] scales_1.3.0                png_0.1-8                  </span></span>
-<span><span class="co">#&gt; [111] corrplot_0.95               ggfun_0.1.6                </span></span>
-<span><span class="co">#&gt; [113] rstudioapi_0.17.1           reshape2_1.4.4             </span></span>
-<span><span class="co">#&gt; [115] checkmate_2.3.2             nlme_3.1-166               </span></span>
-<span><span class="co">#&gt; [117] nloptr_2.1.1                rhdf5_2.49.0               </span></span>
-<span><span class="co">#&gt; [119] cachem_1.1.0                zoo_1.8-12                 </span></span>
-<span><span class="co">#&gt; [121] stringr_1.5.1               parallel_4.4.1             </span></span>
-<span><span class="co">#&gt; [123] vipor_0.4.7                 foreign_0.8-87             </span></span>
-<span><span class="co">#&gt; [125] desc_1.4.3                  pillar_1.9.0               </span></span>
-<span><span class="co">#&gt; [127] grid_4.4.1                  vctrs_0.6.5                </span></span>
-<span><span class="co">#&gt; [129] slam_0.1-54                 BiocSingular_1.21.4        </span></span>
-<span><span class="co">#&gt; [131] beachmat_2.21.8             cluster_2.1.6              </span></span>
-<span><span class="co">#&gt; [133] beeswarm_0.4.0              htmlTable_2.4.3            </span></span>
-<span><span class="co">#&gt; [135] evaluate_1.0.1              mvtnorm_1.3-1              </span></span>
-<span><span class="co">#&gt; [137] cli_3.6.3                   compiler_4.4.1             </span></span>
-<span><span class="co">#&gt; [139] rlang_1.1.4                 crayon_1.5.3               </span></span>
-<span><span class="co">#&gt; [141] tidytext_0.4.2              labeling_0.4.3             </span></span>
-<span><span class="co">#&gt; [143] forcats_1.0.0               mediation_4.5.0            </span></span>
-<span><span class="co">#&gt; [145] plyr_1.8.9                  fs_1.6.4                   </span></span>
-<span><span class="co">#&gt; [147] ggbeeswarm_0.7.2            stringi_1.8.4              </span></span>
-<span><span class="co">#&gt; [149] viridisLite_0.4.2           BiocParallel_1.39.0        </span></span>
-<span><span class="co">#&gt; [151] assertthat_0.2.1            munsell_0.5.1              </span></span>
-<span><span class="co">#&gt; [153] lazyeval_0.2.2              Matrix_1.7-0               </span></span>
-<span><span class="co">#&gt; [155] dir.expiry_1.13.1           sparseMatrixStats_1.17.2   </span></span>
-<span><span class="co">#&gt; [157] Rhdf5lib_1.27.0             highr_0.11                 </span></span>
-<span><span class="co">#&gt; [159] memoise_2.0.1               igraph_2.1.1               </span></span>
-<span><span class="co">#&gt; [161] RcppParallel_5.1.9          bslib_0.8.0                </span></span>
-<span><span class="co">#&gt; [163] ggtree_3.13.1               ape_5.8</span></span></code></pre></div>
+<span><span class="co">#&gt;   [1] fs_1.6.4                    tidyjson_0.3.2             </span></span>
+<span><span class="co">#&gt;   [3] DirichletMultinomial_1.47.2 httr_1.4.7                 </span></span>
+<span><span class="co">#&gt;   [5] RColorBrewer_1.1-3          tools_4.4.1                </span></span>
+<span><span class="co">#&gt;   [7] backports_1.5.0             utf8_1.2.4                 </span></span>
+<span><span class="co">#&gt;   [9] R6_2.5.1                    vegan_2.6-8                </span></span>
+<span><span class="co">#&gt;  [11] HDF5Array_1.33.8            lazyeval_0.2.2             </span></span>
+<span><span class="co">#&gt;  [13] uwot_0.2.2                  mgcv_1.9-1                 </span></span>
+<span><span class="co">#&gt;  [15] rhdf5filters_1.17.0         permute_0.9-7              </span></span>
+<span><span class="co">#&gt;  [17] withr_3.0.1                 gridExtra_2.3              </span></span>
+<span><span class="co">#&gt;  [19] cli_3.6.3                   textshaping_0.4.0          </span></span>
+<span><span class="co">#&gt;  [21] sandwich_3.1-1              labeling_0.4.3             </span></span>
+<span><span class="co">#&gt;  [23] slam_0.1-54                 sass_0.4.9                 </span></span>
+<span><span class="co">#&gt;  [25] mvtnorm_1.3-1               pkgdown_2.1.1              </span></span>
+<span><span class="co">#&gt;  [27] systemfonts_1.1.0           yulab.utils_0.1.7          </span></span>
+<span><span class="co">#&gt;  [29] foreign_0.8-87              decontam_1.25.0            </span></span>
+<span><span class="co">#&gt;  [31] rstudioapi_0.17.1           generics_0.1.3             </span></span>
+<span><span class="co">#&gt;  [33] gridGraphics_0.5-1          crosstalk_1.2.1            </span></span>
+<span><span class="co">#&gt;  [35] car_3.1-3                   rbiom_1.0.3                </span></span>
+<span><span class="co">#&gt;  [37] Matrix_1.7-0                ggbeeswarm_0.7.2           </span></span>
+<span><span class="co">#&gt;  [39] fansi_1.0.6                 DECIPHER_3.1.5             </span></span>
+<span><span class="co">#&gt;  [41] abind_1.4-8                 lifecycle_1.0.4            </span></span>
+<span><span class="co">#&gt;  [43] yaml_2.3.10                 carData_3.0-5              </span></span>
+<span><span class="co">#&gt;  [45] rhdf5_2.49.0                SparseArray_1.5.45         </span></span>
+<span><span class="co">#&gt;  [47] Rtsne_0.17                  grid_4.4.1                 </span></span>
+<span><span class="co">#&gt;  [49] crayon_1.5.3                dir.expiry_1.13.1          </span></span>
+<span><span class="co">#&gt;  [51] lattice_0.22-6              beachmat_2.21.8            </span></span>
+<span><span class="co">#&gt;  [53] cowplot_1.1.3               pillar_1.9.0               </span></span>
+<span><span class="co">#&gt;  [55] boot_1.3-31                 lpSolve_5.6.21             </span></span>
+<span><span class="co">#&gt;  [57] codetools_0.2-20            glue_1.8.0                 </span></span>
+<span><span class="co">#&gt;  [59] ggfun_0.1.7                 data.table_1.16.2          </span></span>
+<span><span class="co">#&gt;  [61] vctrs_0.6.5                 png_0.1-8                  </span></span>
+<span><span class="co">#&gt;  [63] treeio_1.29.1               urltools_1.7.3             </span></span>
+<span><span class="co">#&gt;  [65] gtable_0.3.5                assertthat_0.2.1           </span></span>
+<span><span class="co">#&gt;  [67] cachem_1.1.0                xfun_0.48                  </span></span>
+<span><span class="co">#&gt;  [69] S4Arrays_1.5.11             tidygraph_1.3.1            </span></span>
+<span><span class="co">#&gt;  [71] pheatmap_1.0.12             bluster_1.15.1             </span></span>
+<span><span class="co">#&gt;  [73] nlme_3.1-166                ggtree_3.13.1              </span></span>
+<span><span class="co">#&gt;  [75] filelock_1.0.3              UpSetR_1.4.0               </span></span>
+<span><span class="co">#&gt;  [77] SnowballC_0.7.1             bslib_0.8.0                </span></span>
+<span><span class="co">#&gt;  [79] irlba_2.3.5.1               vipor_0.4.7                </span></span>
+<span><span class="co">#&gt;  [81] rpart_4.1.23                colorspace_2.1-1           </span></span>
+<span><span class="co">#&gt;  [83] DBI_1.2.3                   Hmisc_5.1-3                </span></span>
+<span><span class="co">#&gt;  [85] nnet_7.3-19                 tidyselect_1.2.1           </span></span>
+<span><span class="co">#&gt;  [87] compiler_4.4.1              httr2_1.0.5                </span></span>
+<span><span class="co">#&gt;  [89] htmlTable_2.4.3             BiocNeighbors_1.99.3       </span></span>
+<span><span class="co">#&gt;  [91] basilisk.utils_1.17.3       desc_1.4.3                 </span></span>
+<span><span class="co">#&gt;  [93] DelayedArray_0.31.14        bookdown_0.41              </span></span>
+<span><span class="co">#&gt;  [95] triebeard_0.4.1             checkmate_2.3.2            </span></span>
+<span><span class="co">#&gt;  [97] scales_1.3.0                rappdirs_0.3.3             </span></span>
+<span><span class="co">#&gt;  [99] stringr_1.5.1               digest_0.6.37              </span></span>
+<span><span class="co">#&gt; [101] minqa_1.2.8                 rmarkdown_2.28             </span></span>
+<span><span class="co">#&gt; [103] basilisk_1.17.2             htmltools_0.5.8.1          </span></span>
+<span><span class="co">#&gt; [105] pkgconfig_2.0.3             base64enc_0.1-3            </span></span>
+<span><span class="co">#&gt; [107] lme4_1.1-35.5               sparseMatrixStats_1.17.2   </span></span>
+<span><span class="co">#&gt; [109] highr_0.11                  fastmap_1.2.0              </span></span>
+<span><span class="co">#&gt; [111] rlang_1.1.4                 htmlwidgets_1.6.4          </span></span>
+<span><span class="co">#&gt; [113] UCSC.utils_1.1.0            DelayedMatrixStats_1.27.3  </span></span>
+<span><span class="co">#&gt; [115] farver_2.1.2                jquerylib_0.1.4            </span></span>
+<span><span class="co">#&gt; [117] zoo_1.8-12                  jsonlite_1.8.9             </span></span>
+<span><span class="co">#&gt; [119] BiocParallel_1.39.0         tokenizers_0.3.0           </span></span>
+<span><span class="co">#&gt; [121] BiocSingular_1.21.4         magrittr_2.0.3             </span></span>
+<span><span class="co">#&gt; [123] Formula_1.2-5               GenomeInfoDbData_1.2.13    </span></span>
+<span><span class="co">#&gt; [125] ggplotify_0.1.2             Rhdf5lib_1.27.0            </span></span>
+<span><span class="co">#&gt; [127] munsell_0.5.1               Rcpp_1.0.13                </span></span>
+<span><span class="co">#&gt; [129] ape_5.8                     ggnewscale_0.5.0           </span></span>
+<span><span class="co">#&gt; [131] viridis_0.6.5               stringi_1.8.4              </span></span>
+<span><span class="co">#&gt; [133] zlibbioc_1.51.2             MASS_7.3-61                </span></span>
+<span><span class="co">#&gt; [135] plyr_1.8.9                  mediation_4.5.0            </span></span>
+<span><span class="co">#&gt; [137] parallel_4.4.1              ggrepel_0.9.6              </span></span>
+<span><span class="co">#&gt; [139] forcats_1.0.0               graphlayouts_1.2.0         </span></span>
+<span><span class="co">#&gt; [141] splines_4.4.1               igraph_2.1.1               </span></span>
+<span><span class="co">#&gt; [143] reshape2_1.4.4              ScaledMatrix_1.13.0        </span></span>
+<span><span class="co">#&gt; [145] evaluate_1.0.1              tidytext_0.4.2             </span></span>
+<span><span class="co">#&gt; [147] RcppParallel_5.1.9          BiocManager_1.30.25        </span></span>
+<span><span class="co">#&gt; [149] nloptr_2.1.1                tweenr_2.0.3               </span></span>
+<span><span class="co">#&gt; [151] tidyr_1.3.1                 purrr_1.0.2                </span></span>
+<span><span class="co">#&gt; [153] polyclip_1.10-7             BiocBaseUtils_1.7.3        </span></span>
+<span><span class="co">#&gt; [155] ggforce_0.4.2               rsvd_1.0.5                 </span></span>
+<span><span class="co">#&gt; [157] broom_1.0.7                 tidytree_0.4.6             </span></span>
+<span><span class="co">#&gt; [159] janeaustenr_1.0.0           rstatix_0.7.2              </span></span>
+<span><span class="co">#&gt; [161] viridisLite_0.4.2           ragg_1.3.3                 </span></span>
+<span><span class="co">#&gt; [163] tibble_3.2.1                aplot_0.2.3                </span></span>
+<span><span class="co">#&gt; [165] memoise_2.0.1               beeswarm_0.4.0             </span></span>
+<span><span class="co">#&gt; [167] cluster_2.1.6               corrplot_0.95</span></span></code></pre></div>
 <div id="refs" class="references csl-bib-body hanging-indent" entry-spacing="0">
 <div id="ref-Argelaguet2020" class="csl-entry">
 Argelaguet, Ricard, Damien Arnol, Danila Bredikhin, Yonatan Deloro,
diff --git a/articles/case_study_files/figure-html/abundance_plot-1.png b/articles/case_study_files/figure-html/abundance_plot-1.png
index 7561c1dc6ef5f1f8af6e22b73853020c51f47de9..f57653b06e8369fa17cce31f00e8a8984480c5e1 100644
GIT binary patch
literal 74234
zcmb?^30ThSy8efSWo)lShENM%X^;#hQ8I<ENTqp1npK+T!PpXsuSs)MnoBCpg%m~e
zq&bb6=hFPYU$pi<=j?s<Ip_L!U288Jef@^_eV%)I-gQqx{LGq_Oe-lA${O0)Q&JSl
zQcDVD$>m>{;dfLPm#gAm%dUu>IfZ}YYw~2e1OBz*`dI}N3T2xU`Dc-9`;Z2O@&|=>
z>bSI}UvGnzyocG`{LscfW~AS#e0sZPspP5pHH_~=SD!h>>qRp;nIPl$fiYXaBb2-2
z=ELdz`<`kj+!p;DerZ|%7K2+HZsGf$d2aFf!>@@&*>A{l?T3!zqmwCnVpk0mcg|%r
zO}JP$7j|YCZn^d5(%-&c6J_^GRht)y7ujFD*r@1XzLFCC=~d#oAAdxltY5Q~+ynVy
z4`v@+{NtC$HjCWY_~Vz90``>|KYuB4TC#cRk6%0PcV5i%^S9k4*IO?7@#~-m&t9+k
z{!2&O$?55jZ{8eVyLRniyD8HwlVtrmw&ly0i-?G9cE$hX<In0|>3^J%zpS9B{%!D!
z@%8Hx*4EaKtGUl^ZN7Yb;R3AE%=>%}>eq&R`gBf1Q&YFlhi|f|xTr{0Q}g-DgoIUg
z(Jyw#JoeS1ao$tDvHZ@$RShTVl(W&K!o>Zlw9A(-@87igj$%$-ozkj}d)%Fz&Sbq`
z8AYwKe9}S@k&In<9Itcqiu?rVAu>_2I<FlU9X4+J^C64yo(mU>7sr*pxO_5mZmMTO
zNLHRTJ}+nea`u(xKcCm@XZTd@rYt+%&o7?VH#9JA-MX8$l7WGz{S&3_%5uZaCUY}Z
zx$oY&n=?@;?h4}K;!M2S5|4`-k_`jbvK&=@yLl01@Y{6@3$M61L!Gc2C%%o>8M4yW
zv7LKF9qTGTLr=6Ex`O)_d-ZB*-H}b(w#B@DeGH!$ZZpxp!L`u;h@9K53sq&IVtl$4
z><N6as+p0BNxBi{eYK)A+G7_Nmt3;}&nHj9EgG%PD80J&I5;?XW_Fg(tXn!YEzReX
z7>y=*>eTJiuFNr<GLhP4AvDEI>m>8u&)W?R2W(sY1g)eJw2Ea-O%v_sXDzI3Z1mp!
z`5P0TuKU)lTUVMoIJN}6R?6EdY_F`I_kg317b_`2Sw-c(a`SA5uOY8qRrK#gzfv6E
zzyDeq%rWS>FVKD}f=Zq*J)|~CFG}O1=V8;YkH!a@;&8!?+Wt2Muj=DZSFTtgDKGyd
zcd;RU@csh__%6TwrRMYJwp4>yRxwZQx3^bBN(SG>w+6j?cPd6c{sfn7w1l*@d%Qe1
zEk8eh<(f4zokjk98U^cFsnoOd{rmS1yCm378oO*eB1>=U>Wb(n^yS*V{dio)u(M{7
zUqp@*LzHf1<kGY=$0SRi{anwQ_LJ=!{STXo9zA;0*xdXeGYh|gUx;5o25W>tL)!4*
zV9C>yPNqXGA_+tFs;^JpUCmKqjn`HZDHBD0ceY1uzRY2MS`F)+Mqj*SiHE1Brn$I;
z#G}>AZi)LHdivnO1KU-f-#^?|{yAQ~y|*{|VwiXro|E&vd!-&+a-(mSv#_(Wip<T;
zEiGVVUu!<-d3oCrbLmT$F6o3Vq8#(yYc<?<Lby*rIOpyfH!m-_dojx?uY)-S1Qf(P
zcjH}1{GwMK>p{J9`}UKq#oUUqib<lYHtd#2HD;Al_1Sy*X5NF%!C${>yKLI`)>4q0
z`~38H-$hquA?YaDn8h)C=DqT6{;v-nJlI_x=Bim75VK+(n|3~()#s>PTE4)>J<1YS
zt~}$E4!_<sW>W0(;DL1TX_p7@_4V}`nV2HEm0qzQJSe;CLV$aEy3GyaWzIJj8*Z*j
zGt)cocPKuwoI(jY_v@N9mHtO8M6YFHL5~eIWxuOhLTQ#vQh&d8WroJ1H<wB>tVXD`
zOzW}zn@#ZeU+J7Zdsf10kBaDSgG-l!$ki`hwc*Kn%bh!SveGU-KgThsn0mxzfX}$?
z?D;2qkM~r?9N-$v8Z8rB%Gs}}rNy;>zZ4^vY(@D6VSOX<vuFRjqOQ)kWsBI2jSTC#
zu{X)rrIv3)t$7=kRYuCV-MV!v{|xSts?${#%F2{{+J&jRBEpTV)sJ30TU?9151hg0
zNms|JtjdsWNHgPRj#JG{qjtT!w@y>Nv9WQf;`G#5PeD<Ur<a#kJj-Av6TiW3ny9EK
zXNk`-$@p78i!9sG>D~i00fx0EMn?O#WSew8d}YuO$&>Z%1eaW_^yF~6*zR#(!>ovJ
z-z-JWoYDLI;__je@n|}ozMhFmG2HxbR*b6p*REaF)8jgF<VeCk8=L79?Yert6OH-x
z@pRjbn=h?csPdaMG&FQdpNY^z`s$LSa6cP5v7x#4?UOVzG8!Hk;fnv3s>iLIB7p@}
zo@UlVr3IgH-BWQGTj^qgX5puYYN^H|H8nNeSon`kOpT4{R8}^&f=helceCx<wM&d%
zUtjOMbvgU5il_tR>ouv1q8BcBxw>8m$W@`OT)Wnjs=hBJ<j~>6k!soYwFz3TWo2c&
z7X8XrR#y3yl}QpkZIwrN?AURF&U^L4qaB^vdU{cgj+D@_u%6d9mTc(!6ez4luZU4f
z-tm6X%9{)I{(&dA5?hzU()x^V)6-(~gV#O>ZQ(a?*IsLwF?5%zgNk+7VJ_qOd0*MD
zd~e^r9j%m8^WIwbQse&3TeoDgY^_|j3cQpT$eQj?ryDb`wm`vdHM;dCNWi3nosI1@
z9+p>Fm{xoXW#Z&QW1@#Vc(7W1zCn0?vUsv6MkW1KMn(pe=F4xmr&Q3^&Q8P7Fz(5d
zO*OT(krx7w-Z`9YJ2^Z&I$9BQ{1*OMvWAKGY;0_-^>~ezhPrxiO3ERbNa+Z9fmhdR
zq((Z6t5Z!n+ge(l?NLe79&CPN-Ewc&XqA#-er~Qvlf%4SK&Imj`iZ|^3aSD-7nhij
z^{7O9fyclRVGA!qgCjTTS8x%M!KY~jHMvUhS8`AIt35rjblXu|`QLu~&Dh9D^6J$P
z6tYmG*1WFXnuOf$Y@6if=1bdztfhCJc_=L<MX!oei?bf<*7@@G&SA4|_ei-o)g4BT
zGlP%#e!SnabnDTt+i$&*Xr4s*FMYC4{oF&LkxvIS_3$1xe*N6wxktrEi=|<vN<Q5@
zW@u#l$-C>rs|~PYMhaQ8m4$?alJN1<c@*aL3+<HhU*N~!ulHPC<z}bG#&Q^h-OV|F
ztgMy5D~^nQ0w$+XiY%LrfCkJui?(TKX$cR?QykBDC8;JmcV8}!d%>L4gMwZ~Kn?SY
z-rj3wy`LqIA3q-I4`4Sj_9ELrsCSaFD%e~I>vig1qabF1`MHT~KKmI<vQ%?&PU16a
zQ%$r2M%>)o_;kvsNjl}HUAFMc9y{iEbbiJJaK)!^cEDji3RQ|7<<+9UzVxNW$BIy~
zjeG|6@+j-IFD~CgFEE;0&SD#Mcg+@2`r2)WLr^pHzLq?thhQP-y}7wG&9sX&Fkav&
z{FqbkU0HaH?E&7-&@v2Eupt9&hwtI<?2wa_BhOwj%Qh8n)kCXZ$Y%WOShboQ>Umpt
zcVv<Oksg1G#?rX#ImPCgetVZJV5{dB0{0ji8uIDaO1Hjutv1Y_)y#N^MZl+5bq=7T
z9Ock#s3lj>gzM9od)vecJp842lCO7aphR|;20LX<c5LCh_F1CPm%kdjkoVGS%52m8
zY+%jIpMU;3IW@(Jy^OUj>9ls6ySsZxzT2)Sv^fIn@@H^^J$R?mD6&)IeNnmhH%6Uw
zTH9Tpny{8xFkDX6O2nk&)4n2y4sR`9;1!w_u4oMCs3&)$nmaW#B;+f;yP~T+jPH8$
zu^8o4DSm@`iPNV~KOR{^2|DCFoo4p6WP1|XQXPfmXf+k_>St6^t{)%B7<pKms1t0`
zS>&DTci1cvk31|qoV_weNe=zKY^<j`319)=eOwrC^2YE-JC{bhtHb8K;j_~dQP>8?
z#>QpUY7Q~XLe`{203iV9CN5sQSUUcSh~4a1^;CJ_+*RXkBAEg7nWI>LK|uu`=8Gv3
zQsfmUH`yzl_dQriJ^^S!luowcMpet>_-L|zYLg9lNTUE6bCXWcJP!mBVLv;C;$;5v
z`j-=Sqorq<j@qW|*txUt%F^Y_^-6+{leT<K-^5<dyyMemJa^s33@avK`-}n)t}y(_
zaFuj(pMC0gm6erwH9u`iNIG%i1UmK_{iZAx6eECwc4vPU3IMTRbxcon9M|5x5~Nx3
zH`(rFU|^up-&~Ra6w7<KE$XP<l=eVlW}XS%&5e<nITmG=lao^vf15Y2CQ-+Y{GG{>
zPCXQjrA3BQqg{%r#-TGaGt~ABgSDu8M=bg)fO3*_D>;QG>P_e&(b2on{?!0K${RCn
zDkCoLYCSBH8lcQNX$MO3diKRFthZJy-N)2BPEK3yr_-fT$FSfo&_}Vp(TM|ul%$Gc
zo9bI4sHnJ{hYrc3Bv#<VJpBBwZHPK?XC<3(B29|akb0A1jw45uy1KfS7EJ>+D<&KG
zqc2|@UUJ1~82ICOevg?Rs%XKdPZiyjQ7*X>w3qSmlDJ_$qn1;lp`jdm_7oj5AL%$h
z+s@-qiUM@bIi3DM<g-m3Z+)>zv5?m8eo=uq-?*%)NTb~rDl`<P0}oZwO!Ke$KU2OC
z`(rgU+oRA}QfG$Vv(%<r#DO-9wYgpJapBi4IYy(=tOqhDsO`NS_wL=3yt9(A0=VhE
zUe?U;M|#NZ+rK2lXV0|r;Evh}2&Q_ER;LG|vWBC@u)TNo#am=>4z1><DJE#T^!D~v
z#Ht**-}!?|ILzC(#smBI00hb0-@tvYgK71$ZHKNOlPc1?a#?V?H$n3~Xn=dCya%Vu
zgM9<C?mhW3^rl_d_Vo+^NrJx{TQs!EtoE3j$nmG5D+6;VF*7sc5yc<MK7MN%Jp{Yk
z+ok>J-$JRD*494TxC~IM1lc`RlV84kX^PHDv&uf*v4IcoOwYXk&8_8T&94`cmV?_S
z&}ca4R9|1;eO_}zaqOL$qD2zscQQ*N6E9L61wlXu&weN<pi2+3)Eg_*s+bJO_t!EM
zWZF&dZhzGD`Bo7F#W7vMx}kvETsBggwr-dBrwLos#nw*1Mk#D8Fg(CirU;9H#_rmr
zC@_`wmX;l0I`s2dj?UI-0hG`ii+LQ+oEYn_WTm0j7Mm262A`3MRTi+awER$7DoHnO
zdADNo>Sg!SuNM0W?2Irq|038sxrlN-*|@zRLezt^;KPR!tnU%qiTun)%homi=uzN7
zrGay!yWHZ-W|nG7#w|xd;FTb^bKTfj3YV0KH#0joJrF4y<CAZxsOS@5(=Ragwe;kd
zFJE3JCCT7%k@olX!{eQrne!ojPElv2z#L+X*I3wmM!9QNEf${7kd09c1><Patxb%m
z`6j1k7b<dpgU;N{q;Bc6JLoi^bx)lcxqKKcSUsk0eOPHDL}AvqxuC+`w|-@`N-Jlm
z|6&SP>vlZ~^JY>bsX7(ml5E23H8nLksnmCeJoP}l=pjKtf6Pu~&+{?!Z(6PR>Y6uj
zbmdD8F*K);Q^V<Y)8<b*l#U2fDE+dZt7EEFZO%&-o%~F=lv{kJ%>+K_6aWALlvEn%
zg>`pCm?^ds%K4K|*(gJI8Cdqvtj1foui4*``P{=1yW9@%rMWbBtt!R!L4QMfPg1pN
z`7@FG710U@Q0XGEy`5cLP`r~#_c80KDm>IwUVagKC;;s(7V<$j+6ow46<VNKR~aix
z%o#wSTzjSGF)@VphG3;=7r1kvXr`*#4&_b<XvNG>C@D&kYRSO-!Ie>RdUzBB{tF5T
zwfFbO;sd*Y{and6qh0up4*K)ydE(Mzf!Wyq_~V#acSUKu`up`;wnULf5OyK33T%<r
zq~l!XP~MhE<y7OxBV5s_2$J$W)~Is9_qSzGm^sRypT7tmYdzi@VK+Ul0}i8G_bO&?
zve<zXbj^`7;OY8xuXIrHNyB*-96VNW@j}-vhNBWEPTa)mQQToQI}$iA4G7Qp+iy?N
z^)`GH)N2PbY;c}Rt4z?k(ASVI2V7qTin*SVkrkxje1M<=A%C})tvS~*fkh`NDM`98
zKAjYJq6>WV<c1F8<GEYmnGc^Hr=VbcTZ4c|VcZL>ArXxjg0NiKww&zWc}0?K2E<ni
z3zoHA(ESPkOsL;)6pF^J%*@RG`qZ7+j{fLXXZ3F1y{lUltw1NwFUnzlj@5eyJ|qN{
zJ6z0b58e3S-e~#wXYL0$1+7slpZvv_{<or(b=#^8Z*Ol=F)=QpPkjFhdgsB__X~3(
zny?N)1)6}$<DzW4=~sk0fV_SkZvSX(VIh}oKU?wq{1!+L?%%9_ey@attLY8?ojXse
z*$#<3F3z?div*1rn;yuH(k%4BcU9k8x68Ec!*5uL=N+H|aBtANZ_WJkyVvLKtbX?9
z(%!GY9EU9jgV2Yu`U`i&@u}ynCKX@`PY;+tmb}xFjcT^gVGMKg@;J9|e|oBT$@fpO
z-1HyiHJz}LlKalibTHOe`gO9k5rTWuEe1Sq-8%WWiO-cj@`2H^I$9wZ>#y6;VcJ(p
zO3FE-$oqhn>2Mn@3E$9=Zo$uNas7HpTfQ4B4X@hm%_a87BR{T~`80MbmZh(|w|4|&
zxEO_m=ZP1szXEA<9XN0ibTC(y3z|dr+?3vmt1>SI+EmqLWo0=u?kKw|-@mIBXeWyY
z>P~G)H8}%#Y5usXCetRF8e!O!rKyfCBZW0idVorXrDRc2QSQ+=b^@REYhH%R>rp6!
zDJdqM;`d?!?GE8>#hj}C1tNj8pG8&6mlq2v-+C67xEC+>65z?ww`=4}Uti3fRT~R(
za)RZ1u4-vTfduI`Wo5>9T-9edx%NZTPPt|?C<+_I?<a_jH@~?>4*?BP*lJiA>9Ka}
zp&&K;$=nbLrHYY$b6gk@VL^+$u5RSe2nhLC0IG@XO%#emi#=eTR1*zL|I!)&)yLEr
zC@f2stuA@I^Ni`tHzO-+YfocMhBzSmZr~K2KHkA6mTfCDSlhw2=*H#^ic;z3`m{3O
zUa_`}blfp(@f<5CMA0tXIKCX|CSTqz1^v9RX3IecDd}XLm5K3i2ns5**o{i+ge^a|
z-g3mWua=J{m1^v1NW{JTjFtd7iREjzKBD>`GCFbIhxajc)~xT%ju;u~5mjxCnHi&d
zb{6$z4`x&92W-^!X2q&j&CD3LhV7aNOjcE<_3H=hx1UR~kC{vs>0Q1pIpyfL<|TH6
zZ<e#Y7acijuyXpHu94013%rvft=|SJBje7{^Yx^3>|(@5><u$hN+aWKDu(&(=j7++
z<((a}S-ht0413F_n3JXBHfxjt_;S&R_C$>UEvARy*3lXjUMBH5WRwFml5&HmM@srs
zUDMoiqJnJRym{rXzm|jr+DDSo14Rt0@xUv`uh6BA&V5sF?MR`k_835Gs=#w)gg^?8
ztOEUoY2TIGX!fkoo!wGV(g5F|JQw$mM6v%0G%0!R+=H##wlQwn6drLgED{i$t@G5G
zGX(kW+qW-Ii<i;dA9O_zMT;H+F7*MLHFk;|zKs?Q9g_z9B&DqEM|4rb9MIcjp(-h;
zS~L~oLUmF3NOvp>5aQzD;eiUL-TeAS-uRZKg<q;-l%~E0yCxd9i-tabUhwgwm|^Ca
z=iRm2bju3Vp(I%k<!&T87#PT#-kz4u&M?&cF5ot{q$CI%+F)c;Gb6=j*uRzGk|%L-
z7494t+xq(rhV*6*J$<coN<@EN;?W4$`BT&x-)(9jg+nwCwA(y~KbCP#WTZ~BbhQ3H
z-I#SEl+k8xemcFfrQ;VhotZ@5>?&V-{VPc+nFG!?oR?!NTSTSYUd>P6m@=AG^s35^
zdEf5CQvI2=W=m0-VyG2MOwC09Wh-vmc)OiZxtaX0_|#mf<<%i&Q;J?4vstsb*)dy=
zq__@^ydR!FvxQ6em#KZWLTUWwI(Du~KGhFib^(ZbP}gkQ3j}(^4f=EhI;+n7+^lYO
ztbdIF2L}hVTn{YkMw<a4DBNekzvMtf-sa>GJVc%Z)80#tgv|j<>b2&%Kzgsnj&t|%
ziJYI9pHsUR=b;A?jp>MaG&LWv0K7pCty0yhow3kQ;Gz=)i!7+D%LuAm6j^Sns;Wfs
zL<xRA-d9J&vHKf&h`yd=)+1}yS6gZMFw`lU(R`bOg9E-%46DgKAtB+!$&-bOH?ziT
z^%OHK<FD!KS3W(t92`fK&>9Gt>o;t8j;<T2Q~q32Z3`Ae;gfwQwTc5|z%0zL%k@9M
zIDTt6L)B30`>~=U1M3et;cx4LJwZRLa$K~;6tvE6qQQdc;I$`nkQI7nCz|++3JVGf
zCcc<pi-5223YY`j0H%`eMKq&|OdAW<UAy#vWr?V7Kh?Dhuq405+I7{+3@ac{!gPsh
zQd3vw>(Y7fUUo9$Ny*}LCEtVkr0pfNVsk$2>{mA^pKRgN;)!mwusa&<A0?x{gmQ_d
zIJ&=QS%y~U{=sn0!hr*e1T+3YFJ~+M4XnzK*b3-n3cCk82pP?q9ambqY?){QhxIsM
z22{uH>bOAASQmx}wUp~mA!9rP!O4C9ev3t>=cwX8Xyf>8y#5a<jZQ4~;dqGoO-*XT
zv!iY`W4|n0rVCw~yx?olLHBM-G-^G)^p{`4@pOrH!YLPP+cd3MVg^}`kV`1glAydL
zkT%X&30Mt>Y~WFi-oUNMzJ2?fZ?Q(Ssi~egA}D0e&rLi3`RCq`m*Gz_2^fb!1lO;>
z<L~FU+rq+vygt++x6I60JCjEt=%(F$$xT)116g`Ap`^LCb#y#eNYLye9XFRr;rMy~
z!!lkxYO<D=DOjGGf>~sLca@0{;S7MmIb$x^aj_vYIA@sp`-fz+M@oe{*L({V^P<uS
zS$;es@k;PvZjrgu$>=RFh6<*O-z*tAG&j49(ri$yQh%lW&x||UUgzla)%scR-M#%}
zPVnnMV0o1$7OCQkOUDEBIk6P=c7G711N5?;dlCfIajdVdTK<s-m4HAz77*l%ii)*M
zl=QIh=pi3IoK-jCx5z9hkw9}20mnOx$5o>=4>2dwOU=IKo7R>sTL`m+MO1Zr#kxf6
zG2Q!1OKdzU4ufE?T)DDLFD_G4s@~&jhFof~bDCVPGvj&CWg<sq*-dvt)-FZAlSbp5
z?)al!P@=!TC6-zlr>15%HChVo!A;90I;f?kg-Tnpd@Zj)e@{;oWF^Sh`Iq(zRt;<l
zdhP3kSN9bL%3+IszufHGSTb%@KJAjAQ$<`+-#R9ShH@BtHdxvy@-b0@UVPJlFY0;_
z+OE||2aS%dud%-an&7qaM~@x_0lLd8?@U%3XfU=&QDpk4fB#ZMrUotA!{0v!&q5|z
zUh;zQY+|n8$rC3Cu>bs7K4G!p_3JNkO-?`D8bJ(_zS^XvoS!@^4&#=}3OqV|g9+{G
zt7WBGy*lyh%9Sra3kz6;nO|0Pd${>P-W8*<{(~EHzqGangEs5brx+3DFC8Ynt@ZPx
z?I%U4rxniJ;Ahx&*t7(0*;t&ztXM!m0GoQUVH4ppGO5NI37QQeLJXFlHJ+bQ5_r}*
z8`gY-y|AJ7^4e7~o1Un&95a$nl3A?(JB9Kh4Fca5i)_W(d~XjAR+?g(sZPMsMjIAc
z*^>C>NmVEXxKC~#o;X1Zn}ZN2>Co%09`vl%%*2`$qXQU81PO}jZsX|ed0{hQfIAgi
z&cF;!?Y(L5=Y1MiueRB!_M3U$UY>bx2(o878ebGdA#J>KIFgZg)~v$9!jqGe-iZuW
z%970<qa($_V)X8+m?+dMv4R@aek=<#y8Pne3(U*{LJ@d$o;+%?ct1;*F7=*ulBmEI
zkWZqVj#@0Q<fbKd1aO#HY3Aj+*EuZ-#Yf%TNB*{d<Hn5%Ly(xX0V}6k?(>iyt1c9}
z`$eM3SYtf?V7;Xf7A<5Vru&T|GM~-hKeXZbx3yg*>MCK;(5lc%W6?M}(^ChJ3;&rC
zzPoR$If(kTYtaDDi5ed_=r?94pcC`H`kVJ5qi292t-pVKy8GE{uCAEmMmp6F%1++R
zXt=M0f$}$6o^)e7pV}4rWFZSccXf3o%o6;JlaEgZ`XQc!Z;c|}CFN4X@{Gaj8!4}I
z{;WPhVYmf-j9W4BEIkp_8tp!(MG%w##tY}3J>vHE_EqQN&-(fKQE5<%tj4dM7Z>-W
zrlzKPyMR&4XKaL6$=APn&!2QfEvU+(;^I{(EtLsjI>9+D8+^sY`q|Ejk)i`ukEe`%
z#SSo!LQuXd(DvEf>~v?1d8^T}8~iLF!PO8Ch<T!eN>U3H58Vpd?_M+&j^ZYX^<#qW
z+p2oci18=_9xGaY;G_r!pofNyj+#z>`_`sV?_q46k9B_<$cb~^QdK1-CDdtlh=+iy
z5@<S~I==fmyA6tz&_q4C4;WqqiQP_{`lvQ1g~fMe8+I<(1az3v=oGH@#9+%!?N#r}
z%F7G$uph!8DEn?rx1UW5u<F<hk^Qr!gK;%PjK__36DmYwgq(X|Y2z6WyqIUtp7kH$
z$rd0g=byjr<h}Cl&l9<Ox~(XG+&j2};^_3YO_48#X<%p6=_Qml*5q$bs&if&gLnO9
zFNI5miZTc$w^*YqGcq!U!o?(e;)2kadx+o7uKH9H#&zq0asR$Tax^+w$ilN-7nly4
zh7WxPP2&Zj0eR|b_fp#<u~tN((^YtG?_>h5jJyqZ_w}=#!t;`#@h<Awv*W&ymZ>z*
z*rf##fAkkVWjCH~hnRH{VA0Lr-=EqANXaB<34f#aoH>ge#4p(3#9mXi>4Vo_k$=d%
zr;7W|ojWu-A-lcx1uxH}A+4cN`CMNshAGa%B<k)Giu-Ru_RIzij-G*~ODHd%{S%9e
z@LIC=$v)we+U*I+Ul;lkY3`7g+fbij6tQDr>gedeVk(_zaxfffsOHHWr9)T|xpVh!
zBmiD*wnKJ&=*e^E&Sj<h-%opphJ+RDj+6p4D;ZKcQP5clz@}$sXA24n@@kj-k(!c%
z1$>gY<B-LPnE>2NR2!fD8qP%d2(j^0TYIM+*MJ7g3Hy1^o;}uI?rd}q2#7`3K!LkJ
z%6G>nZ*@OIxNf)yW$YiKbilZ*c%`pIIA~q6p&ZgBpjzA;+E{j3S$Hga6!vLUJId}m
zC%eRb#JpERR@ReXYjD9P3;X^(P4!jFG>)SYJY*J%1b^Y=;GjW7mxFhm2)!N*QWU!<
z*|L%BDJby~00vY2=>w)?-7+rN4=@oOMxjhI9<sk&eedBzPhVeOJlW^y(5*v|u}M?`
z)T0_5F&O!Wa0EOK+kSg-bn5F)yUBJhB5k$g-ZumKJ<$|)vrR&Ne0*H=+&MxDK!3wP
zjfut!*IEwf<L1qqi3SaaTZ>d_*8QoSgsEcV6TmSw{E^3O^sD3=l=`8eA&u^V##?Wp
zx_$rpXI>xK)v0%pTM1CdUGZ(<%1`R=K)M1q+bIy>1>fN-WRdIvK?jF<+t!Vy+i2ur
ziw6km7=3#Gepj(sn#DjY;xkVQi+SP*7X+MNTJXhQkC-`Jay}E)anccVs4Ag&VBRhu
z)Qa^kb%{C`@9S+}4p-apUQAL<9e<RQOz<Mg%HZQ4pJjTGbW;`aX)o>Iiu^LW%Oj+;
zY7O*OqOskEVf^}+^dt%;)&8NHdZlH*KAR#Rq?y;iUgVF2KP|a)AlhF~Ck%hzarsvi
z^D^iIGH_3wu54>oC0QrO$~ywP6R#vLpWb^HSd97WlqS>2^%WG?4Lw<GXBno<7QGH8
zd=0`~-^qlv*Ll82sB&K<i3vyLwZ6GcVdnEKvXSRgH6*2_`AskhO-ZbFh!(os;P_^)
z%7t8P^HWWwTH>~S|8l;&;B-^mB@?y3|6cRHeLpXh{qoNUmgTv`BU`UDIJ(Fw@-tHe
z?`oI+{C<kY`XfJ6TEnHWMjmD2izvhI7T)ZC3V+=v@iyH3Of-v1?dJ9MBZ8mTTGqpa
z`T1c5|F=vSB*66xL4^PG6BnH9|ETN{h3?0dn9W>geHS@oO)#8bz)y>I@X3dBuCI(w
zUEq4Z#PK~bC%%6@3WIn3wUEC_8mgKMe4@%|g^SqbQ@shrrf6Gb;yf3S_VI#NhPEMg
z#6xNwL{LZscw;e%7vC&7wQx0>22EKpAeB+j(0%v~1H=Q5#$f+S0om4OSS1j|7ZzYG
z3T1as{1a5$DQG-mS7Ux8)jU5avc^#34hW;fgPKX3n~yJXy;JsAet->bNCm`Lh}PoR
zIx;ZZ4H`2d%EK-Iw<w98KVP}UFykq9Y!t354TL6M9T65HWPvfMtx$fG^U<fe>o1+F
z^~U=XT~C+uP@3cU7nUzYP?|6s7Ka(hMT-_q!8g^=(@Q%ucyCug$<3v!t|A<Om_c`|
zt8kUgK&G5%%?=*OOcgzG*{ZcV&5P9WY2Uvd0NgOl0(^Qv!6njMLQ8)G^guQmo(i)*
z#Zihx5e7f&SsZ#73G)fK<r;1R%(RRvE@Q{8U17GJRZuCww5w2msR$<?oU1d;)&a*r
zh)MzmR~v4GTnkY+M%XR{L=!+rAMdT%i5Ei<V~ho)4ier%NjEk%l?KXBGVPKgpwm~2
z*vAtCaVW-7;{J!NyF$I((M>vMh(?1{R;k;CqY*0^IX^Tss2FBVJ_j!K?G%8JUOD(B
z2jrAnCPGT&#*G`bS#~y0;}U`<&d~w#5MA31x33HRn>06&$Y?AZF*-EyLgtr9=o}U}
zbVz%mzhOgM4)7|1j%i5l41I=(1@C+}WSwqY5cyUp88OI~krWD{qx%9c=iH3|8jeI*
z==6h4Pf`EctcoS}FGLSs6S18h0)DhX+a4{j;+e=QtRnBNO>hbcX54&0lO4({mQ5wR
z{K!X-{zh8kC+`D!sR$ML2$(#l=GW9HkX(SO{iKH57{MNF=Y6yxgAgPR9!#oy=5PU*
zv584Je1P2@$yf7#M;55(J&S`hvFgEi0F1)1-bhHHW`=keSy@@6LsTNaVm8pI0$^DQ
zVS~h3AT+-1^+V3#1EeexE-^8=h$WDObqmEr27QTZ&}>6T08&nRtB}pZYA2N+eMu$T
zzCnKQ)##(7=;&x$mJH9wk1vm3PXqh`ROpV1%aX!3C+Su1QGWv%pbT-w`N0GCsHiA3
zkqB(gk8tE*krBfS+f{$x2GM0gmMn`1{n=_1>d^{u7GHWbKtj|A_>&^4t$<gg3qkrz
zg6YO**r-e=qJjOaHR*fA$v5eE(A(EXfCSc)PRqM{x+Ot3@?2RAUp9j6@FE!}aO4Q-
z$u3NMboYy|X9jZ^kf7iNI_!Z=x&NaEd@}TC57Y<_kPWzn`t>Pf7p`2rS{h2SpT9pu
z;J>K?<SHg=KZUj-aQHA&_2hq29+=4+2N0>m3V;PbC=@*e5u->b8;X!imN76#1NL^>
z_}|IYvRJipr4*nOu_uXAaxEj|0gJF|ll^oIvWKkg_46dF0Axl;K9WB}8<$|^Ny8X*
z2lYj`4DWRIM%yn7L7e72P%AvCGT{<WKj_`QbB6<tN0Z%zf4l?-vF(WJMl?ArbX`>1
zrHaJq|0x;B(V>5l7$7uNj-y9aA<gn0vFQCAPjaKsXdVcR{?H6$&COrIQo7>j3y1~+
zg3B)mN#V+uugFL~p96J(gI1ej6da$zMI$g>cy5S+CS_^)%`Ptf&_wUsJF9Bb%%ZR^
znB|S+3Ti6-ZCFD5riGCeZT;Mkp4Hde%L~t*CPgHTk1G~(%GbSM($Wg7J}XL+<fys1
zIq~&!hp;+mbo2gttEMD)_Kcf1M}X~gcZCJMI<$*Q?NrzZ{X<jTBKd4e9b8z}_R)|+
zNGn_fs_54y?(^2eA|ea*D2d;M;(?M34Ll&e`=u#H1RtFTJt^*7X|ldEAQP#?T+R<0
zr2~)JeR@dT<>L77L6F?JPwWDu!w2Ojg!+@!J6NI(DU_rR4`>yAsH_~zB7q|hI{g`H
za3rJ!6j~`!QD^FA7IyQv`W;<7*)zsXV{vNPgg7IswAtcIV-zx8^pKDcYQjFTJ94zV
zhF25g6%Y^0V5gGZwN21cf{1Nc-n)-};I`1xRlTD{o3N8mG5FL{e*gygbRPQ(6gd&5
zoLeMGHTc<grtWxD-t<|Ac-;)AQU<vb(Ho+|DtF+L1(^`<i3CpliYhOYY|zku!~yQ<
z6v7T;=r-h<^BQ~*ghn9cbXI3tYAWL)9;LdBrauFYm_v<Bh|h3-izNo%o6qWyGR$yW
zTU(U$Xw<N;lbwNHfq{Wwo-*0?c9pRzbyr<LkZHL*1{)><JgzD&Gq3X$BJbKYlUbM{
z+2Hk_sx=dP3xE}_@Lu4`L!yLz+IpfN@rajK?=>P#*QzwgoeIuqKQnZOCeXKr8Ub^b
z`0%9vH#av6u8Lmjk-p?0l;CBz-n3Yw)zf!x6fOqe1TyvBMm%dQ7^3&K#=~PH0bFcS
z+2VlbN=uFJr&>h8F{5^&*;sWr3^pGpV(VySrW~y)q2(KHM~wTR^Ihf)grrq(dCJ60
zk(Vy`i5SZyCIreycpa^{#-{z~Oz)kIs(Z)#Jwjj%O7a^vR)B2m4w$IX3MA6@^u$26
z;n7#GUcoOxnm!GZ%_Spo16L)nBM-69b%*>wgJ@+?mJlvcZb1GKMX=C0T}pi4v?|(w
zEJTV#xNWUDUHzWW&w)hrz^|;78|K~Hgg#y{7&;^*)NrjA)E0S4+xRM{G2+S&ex4Sv
zcadpLP4?EV#DoOp*yYUG@j4W0X?P{*Hc8N6Ag;WV-<pAEMHCUzDqyD|o6pvH!H1WI
zT=FBlRq_hdU6pCh2>c-@jeV&eZxJ2pl%9!brn9rN6mq{<J)s15Nr_8hv!;R$N&?ok
zcj?Ua$DpB@_17y9t&sU2;+?HIRj4}BajHTjGe&HCsOZ{$^H`o7u<Ge>Hg#Uk;jYF&
zxfS5M%iy}>OZ*={eu|<*D+3fWYU<ftB*e8^OG}GZhE@y1hhP1^vr@8w4g#8Z{Rt^b
zn;Cc(is$Rwz)HXW{(D7|o*WYBM6gD_lV1BuKMJ{863~YvdxLlkV8ff07~2R}wrH{o
ztEiBMb`t?uLWBaW@=8Pu(Hw}@(_S!ZIPmm8a%Dh;0}sDP-2Po^G*dzE7GTv!tP}YB
zQuyaPsZLAl#_A!02xQiC=FB`qX=?lO<tb2#4qgDdrI&6+_)~Pk2rPFZDLg;t{R~bh
z2;o?FD9w$Hlz1>c;DO9hxKdHLJ3`W5T+V6h?6kmxD-hyZsivkzVuGllBzX#-j6f6O
zpCAScl=OWuA>V`51M1<83ax|r5K=PuY`<M6e~)O5(25CYMhc7zq8aSqBujgwdH~nf
zZ`~S=R(BrOZ74JllAZsG7s_@-+61BLkA;Qj@uV)ot;?7HcQl<}DGv4{;n&C^lN2tb
zU&t(C03zkU5$C4{L<MG@*5T#CUa=b~<R^uMyh`B1o?zFbWl+XBuuiyDGvwh~?+utJ
zJ;TBQv;a+863!D4WF>-#=oz#?yK#jZaS7}b(JufvJfMGYTzvkas7M?e<M8!nM*`di
z2d{&k>);C57y@&0TsExv{){LuKIG+{1|%CE8{;mUM4zdgnH+HuG9Cy+tPwi>)qqxL
z#qfR1Zd55Qq;EIb^ff|xbj_Ndv$?NMBn9~7!-9su1WTPCHxZ0R0Nb)#+l!6hZ{2T&
zL0=ZFkU-FI!oCX^E+oiq1VXaP8yXucH0iaUe0uf&2QidREfc4(M_%qz7CZ8D#wbYo
z{~Y+>lv`lVf)xR33bmUl#{xdT`7xRF->^`IA3}%Y%+}dwi3iwIIvhCqR#W6RFED3H
z(1R<>2X<a6w_=FO>(Q_LLBlDO;lCu3{}HVn`VpRft=^S-UnzQPyaxfP%*%U!u6+uI
ztB?dDyWl<jLd;PT$)e<*{y|Cy<(<tAX;xYmf6R#-f|0R&A!v`(G>J!1L`z?o$A147
z%BAN>jDkpVm(*jAcEK)mc)fYihgF1wi%O>FY_D4L*)r6OAmpHP$gD{~^ulJ}wS2jg
zZKo_U!>6tw)E|XVq4jY7F0rdZ%|?HXWo_td%2orsvHhOAh9<o;$*@TknG^fwA;&W-
zg#PP9*urxcz38SDGyT2AUFs|jL@XtNIJF?ng&3zmk`U~UOlZVlK|_n))Oi8fr1p<q
zUevytgfjpj7*jZ)FoMol4h-3jIz^ZrM!T@+D`D*xwuFU*Xd_a9xKBBf6C|$<c%=aH
zdC>b`lSnK5f6fO3Mq+*o{$r#%Mu|KZ!8)~Xlapk24O#AeThfhM?;f3OTL<1mBBrFS
zp=p7bcx#Ej*h5qw_vfql7Vh?km0bx%os{%KL{_nRNrKvZppnQ7Jq;F3d_@xz6V+%h
ztTaSnrVt4o1>xHA9yF8as@j-&Az}io>M_uClB5G4CUk{11OEbyi3pD*qk!3|ShZ{e
zDQOiL|317r&+rJzfD?*ECE|=@1zd3-P!r{1mG`B8Z^r+Mw;k{`blJ4jKhVP?ln5JI
zT1qoM?MI~tR7s$c^biE!L6ajej7B2SSU{B_v~@6Wm6RE(a!*zPZHf3k+?J@D(I^xU
zuPcak?XC|j3`<`<%cLq+g$sGGgjST!YJh4!oAGNkwH>|t&cVb$-bM4%lc&V%ZERE}
z4+}7zP24Yh*%~J8nQudR-)G&xe+UNz^pOJ%VWlm*bsWlTHHJmn;pz~`j3fxv9xn#B
z9}2UcaN4{fSgblABv4LeQIf$+s2rWg4O#5JJtfKpO5f~g`GvfsaXfgW;ghsWPoQ%V
z#!e+dX?IOR*oHmI;Y1z6!jru(`FT=gA0p~_+zOzh#H1hk_Hf$~680wo?=Epa_Tee^
z%?rESQG}rF03n;<*%Y%L&s-DCuW7?4rQ&J8la&MChv<`M*Xr~S#0xCwzY;IaPE{n@
zfkQl;oSbMw7Jsp@FEam;g*ApOee-ul^~lG@W&YcHVJZ!mE9*k8|F?}BchQ&*T@T7P
zLbDb*b&8#rw<4rrnJ7_OzkkV_BB-!=^<7(Gq#(Hiwya-&RjR%alWV(aOF37^{xdSA
zb8sAir3h>fw)g*uRi$VG1yDJl)6-4iO<>IcC~HP4CCApx<hg9=f#c~T?2uJ?Wxct*
zB$Vc}bo?#>NN({+J8NSO3aMib+12Y=7PP-hk3$#hABXRgA|Idd@})bSzJr$M^rC?)
zIjk3~=*q#4zCL~PUKo2nhO*=W?&Q~JxY6lU8te+XyH>m`9m-KOQ}BzC``!Pohq?Lw
z7`w^>qm*cIPER;s;S>jAh9&!WK$)MPKS6zBye(77{!ULlu^rDAaPW#Dp?7j_h0?y$
zk|qUWs_7BP5)dC5Y6tPxK^m%Xi%I5vinKBek&OZUc%5C#et2O9*Yc}7mz+4j5UFgK
zGgaOZLov2UV*4?JrF;Oyqio(e4~5yCiZP!<#byl5g6uXnHV`@LIrptRGH-7`H`!_X
zzURO5`_kPuF)%Nb@IeWZMKVJ^cbkCmyOwW=^RWR2p#Enc>mgZt;AaS;1gzu=q}u7F
zoAt2KkZyfRiVn=pZa_}Wk&72ER+8C2@BO*DxD4Us4}Jjw1_vj85Q&VMA#7PP`av*n
zz7Z4?2pDGvvZiB0rZ+kbH80@8*C)SxH_1le|4?ZVb;(dtzPzn%20RVorQDx2AE7Hf
zS-pBSFUK_`+euGBmWyQn@^e}{iu}1~-F3-vSRj`L1O5TsU#{GA$2+i&%L3j|-uPOn
zqMTZ2E4z`3NSKz6kYXhv8pK3FOZYeA4m~kLMx{ZzJ$PhemB>%MoFY%aKA?J|g(|KH
z;-9s5epQt$crLMGAY~Dc>U+`oiS?oI6`%;i&%}&JVWP!=>MY<h$|zE0%1^H(|C|Lw
zPKAWju_sAx8JQWhHn;ot?-LB?R<#ly6B!d3#bSv9ASWG#Q#Lvr!)FZ{Ru{==D2SDe
z`g+Z*xVShJrf{Sdh+YULNjM@fZbh7$un_8<&jGE6`F7Y;rJaEed*Ia*E@fa~KoZD%
zR5QY{=t$!8&6_uDBD(IFM+Gv$*Xw7bm?G~xF$?Ir$2=M{rQk3+H7_-s@83Vb%m6Y9
zeS%~cLAbg;&KEZ}oRLvzpJkb+SfBpI_K%R=ze@-FzO<Z8Bvi8C-EWKure&rXsk`)|
zwG;OzY##WuHDUKf?-wQKw4#rfuV;C^gKvo~C+*JG(h`ZCe2fyj0>(uH3=V!>rbWk(
z-FTR#rWY^Uml>CtJ+0Wz!ooi-N_z}9jgU?RsyO%VeUYaJM#j$1FGrF$Xvaxca#xbs
zE!gKIkeiX2sUP4C5f$Z%#DLKA^Z<{^a|Is`gG<H#fvIv^6EqzHNkEzD_<XFbD6};W
zY(Wg)N@4-Pztc3KCu7QS3X{+zp$hqj3+Y8X8|$whe*N{?LKfhule>q9Myl@OHTTzj
zM}7!vNJ&de_h@0VRpX%Vw((u8tbh;s&ojZX3W|%9r#~TU6pQghX)cLg^m~sUxs|I{
zkxYE-q5e+fsc=uux~_4qe33^QLHG`6ekyse`9N<~*H>!)HjDbFK>f1m%oW1o3q${o
zZaWqLC}o$(&oO=qMcmnO!S?Z6zi46R{}9LGpB;g6UFw&gmrQ96`E>#Jy{7&4zxZJ3
z;JbUiuOv#dTudtqQ`%z6^*<M$#DDx)BN0SC_`_50zOz;}x-3NIC@7R_;h&e|$SIiJ
zK6!V~(^}2O0|`&XC=$gf3&Y$LzxBItc`3(*nhL^Ht6O(}H`kPH$mQ1u^-reRFHvo{
znEjDJ)7LJJTtDq3%Kz?G{^!r=QZ52)%WgeYB#9U^#g7;P`~%?zw5hT(XfK@jHQ26d
zWqq}7fDaKt5r&bvc(KP!MLCd(&y-vyzP_1p@bGw*YD*(L0TDZLYhAcxL!KDo2>Y2_
z_WKrk9(2p62A8tC<(tZws6|=%Y2e!zUel#&xb3VgEP;3T<ebJC0KczW2hnya>dA8L
zvkPzaIw$<6%5@h6hoZdk+x36ZsK2j_>T~w?Ss41_Ld8J%FdRtQy3^l6T2-}waC1UU
zG<^3cguOBS>JBrXqyR7oU%L&_dTU?TV8s2;VR=eON<MyYI=Inx<UC+3lG`T<ImNhS
zbpMbU?}gC%dSw@8Aw>wpWSEa}<Hj(!_cK2Ibpe2-m|#DV+-MyIxddG?8WGAK0w#Ty
z^RHc4sLJag+mTZRjE!Fs$sN;v?HwK6D4IlQLY>kN*ymeF62r?@Zwf;20b_MNuj&lS
zN!^AAqzC!nVDabavkI3lKP3(o=DS2m=>>to6h$a;(2yLchRR3c7F<O;X}C^=y8IL2
zQ-Q!oV5Wi4a7d3akgvLZdg}s!@Hh~DA<O_Z5~)@UK#@6LB1%EiimDP!iT~R|m5oIX
zAPP8dtponnZuODjVPZsJ{-qlRHTsAQiaG9s4`ZT?8#W;RP^IaDZQk89KbMR?83Js3
zJl(uc3X(fMU_E9-Ai`CIcYqE=!B04BJsOUIJ`a#Fr1WJ`_Q|}*gVX9B#RK)J7gwxU
zL4vi&Lz6HZl$VIaL}}zN$m}N0XNa?I%61P7jDx}qaV83X1_Vn+>u`idEV#$VtNXh=
z<WwU{(_g#XaE<=4dSAz4W}0)_f!R~sd%p-h5gA9&$i^}f1H%vEg5iK=Err#E-odJJ
z^!W6r1I1_QM`uSbHrS37dUgVZVt!Z`*zhY!hnpATx?uMvv~^E4nYbW~Q;tVN#76)T
zakN>}Z=emWtTb{gLUM9)WyV1Ed=`mA!2Ci+HmYj+RakWNIYb{wb``!t^c2Y+-qT~c
zw2>4K@^z^i%Zn7{xo?1^X<IAEu?qOPGT>-^MdX|cS{W$BZv8V4w@T{hL@ZmoRhzM2
zmYfy==AkQXg4Q6DqLg1<4VR?<%WNu47&5#~GJYuP2IVZsqnDLQ5$Gb-H=$pQq<RG2
zV<awGwofxb0kT*^^zxbC|GFTr%%F+@1TYJ(?njPY95gik2pdA|bTU&WRtMhCLnGZQ
zCv_mLXAhu^T=LbAhm(xp6VZ$}Rf*DZIzfA&Ub^8KjB%2T8Fu`60RaKT(l5eqa}ln%
zyJgiv?T@_|x%f*#78bLNp6&U(8NcA5q{H|=Du~z9)5VphP2+x;I(Yo}F(*5_1OPrl
z?UBGD*iIjtnwt7oXdPQ<G_PeM1uN6EikZC&HPg529Y219%E8Y5q0GD|CRvo;zH+e*
zpirzvgiK_H@fAGXweQ>7+QPACx)BbF!IQxrmywVNf@%E~b2G`^QL6(M*`zGw6y$O%
zD`kk%si5Eu%~-?YTzh5`B)IrQ6x;}S`eZy3s;0)sC%DpN5~2EhsX2tMbKAFXC%#gH
zH56w;;4nf#!>gDT!_!C78n5!j3X9N>a>pqT7j|cJB#HOZ*P|{03@OwWY9^3UMvg5&
z%tsb<JhT|Y`*E4$KIhonKtRIa^eat~L^dw}90ILl$t!4hiJsIo%!2Yba3LWPuZ0WB
zDt#0(fMTs)!hjl47_cLxWo3zMtT5(()J_q!l#U8->5-l@_UA&gmWu%^C2$UTBu<p&
z2F#_BfnkHx_6=l493l#GzCML~%N#JLJlcGV;k(re(WfHp0t>MkaWD&cKA2AGHe*5N
zz8ZYQ1F=yg4@r&=KVAkE(F3*}=rji#TSp|G4+_UCDhDqwISQm~6;DPG&{fZKUjs4>
zM%XxNJvm?oT6S!?1*P-EPnhATw*%@ZU=h;+%?HHjBF;0m6oJiP*zF)~BM$(0Ng@jL
zZ?@ke`y#3sL1V<?M`-^U&KXb~n11rFP@C37!JoCg(_W;u!xSUOXkf&buuDv6M4;2?
z0r->SasZKWutv1_h9$OFf9kBVSOvPX*YFAvG?ot5A`{erg~ThS<4v%E>XG1RS$oe~
z`#q?-&$;*SFMHnCiO)es%{14_{%8}7#K=r1pWbO4F_DmN0QVKC)83=&Fsp)4u3S2E
z{h@`XJQ&sT<x6P^D@FX)|6dH^!jwh{dS#z(b7GczL&}S}svqc`5_#{RL#=;6a4kQR
zLXMoDGPE>LdQm8Ajc)!sT(|%gB<>W8$LfFmJ+iM+HhzuDVn^$|75@%B8ZGDXlRMjV
zH}-qUrBJjb{Qke9Q$F3Vi@V=q%O`0TG9g6rwIbpo&L*jXGe$@p=#jMEL%7YZoZQ@*
zEh+FlbdXGw)yM9idM`ZpWpFe00S-!ep=kEw3M=>_83A?4=v^}~h(){VNqLs$hF%I-
zCjox!{P1CHvKia}!ppvo@AGOu$~o998k_Ez91?O0?Ul%Q@HvT*K_)*j7DC2Q=7znp
zrMcSpm|P;m!&6#PPP<%F`vQ|O(smppDRO>4-a^n*cQ3Dx+Psn2H8|`_=3$bT%b|4(
z(66Go=b$wAb%UZ=hoEhkSt!SJ!l+mFOzv><p7!>3(&xz$TNqtXLPyVN5kz&5c`1y^
zZpdd9*vFU3E#O<y%CL<6<`@LN-}3d=Dbx%kr^`z;Dk9)3l0Y&qZ{kz2MY`>2X}2cu
zamZa|^`qxR?Hq$FM<S!tE~uOyHP^CQl%ZU%bs(W#cq)3Bus}2-(F!4@Y!CPo=)b+!
z*Zz36yctN#b|bX}hfwXL*W!u-GsZ!9vnq66%fU;&M<GX$ISQ>Jzfz!vI9-2|A%#g{
z{q2jB{uCju!^6fHzf!JT!1UKyt7TsMy1DJ7>6SltX_z}Qka3Z2ifd4eg8{yl49Jp;
zRaVhgQ$shUMu6E7KWGjis4f@&#u9`Y<H)#I&*<Lo*XWqQv3qv~7#0a%kPs=!nPLX)
zIZBI?%SRj~fP}Fr`dR@pNZ`*T7W8ndz?B1?Bu|3DPb{5$ZQgDYThrDK2l61^5)jL$
z%T-{PQ3RvG64b?bG62aaOyjx-O4;_A-~>qOYggL$*9(yI$uO?+2(l>x?1XR5%-A{&
zG(#pQ0a?r^k-25OC?Y*WdXFpOik3RxxW!<1b>-4{kNM*^*4DLs*^?c<tPxeaEP3$U
z5I4O@ZWEBB@1cNE%WbL-B1BU#Np&hQ?Ji34E=YqD<uS*M99o2hTay?lWFrgB730aK
zG2Pz@8YP?y2-ru^>g9rCh_zMLZn#s1;4BnMG7o~#L>C61D&U?)0L+V_gUMho^{2G1
z{rK(e+Y@9+89~}I9PLIz$Yf*;<N4Ki%Diw0i4RATU(veWi0Z+4-Q6?O!Jl0U5s^Sd
zp~*HklHt>c&SdbN{nqM=sWPZ3VWuCsli3d+R>b9j6+A|TCnwGM8VdFS^8(&WSy`px
zBf@~adp`1HS76447eSx)wzdyVP1&*oubvOA;rHy}Z06?X_GtUo<KppDw>`z@``llk
zaNIm5Qq2x^0I$%gm>D98E22f=*a0()bCHvlFhhJNz+uh`)<y9bl??KKiExlSfVQ~+
zl*XRnxB?kwsc8U#iVQ20s8<M={2VR22%b<0DS}2|GC30p1C;1+QC7mp1`CaroV0HD
zkn6SoW<UD1l(birGL1=>?wdCk1Ci-rY>vo#+&!@L7a%^sE8=+EG&uYj{n_*O*w?z)
z8pH?AvCABMyNZZz2<<#W<5#H7{?@XJpo{OvotPa00XT)nM#qZCn(F2vNA1wB9`q#=
z3%M9X0xIBgYegia#dc(V$-AzyG(*mquX?8+)(nX!WAIOZu6ZgIxC}2-RSKH{wMkcJ
zVtl*`6CP0{%!oTC$~hQ}B$=EiU)_oTsvfhagG-<H0pSyq8eA}@_G3W-*$BXlHMjGi
zoYC6`({A7f74M{@U+;sZ9F5om(qJaK7&oTsfF968kcUxx+fX1tvLVEI#cY|(vJ8l`
zo>X!VvV9mO#qdl7P%mzc+i8MmKPWNiBl2%mXaH%Y7~KxXI#$4;WTnWWL*wnau441o
zE#qRz5ndQKx;kNal5^mQk?Eb2GP2R~+w;dAdezeUUe%e_=_l9p;Y6@fOgTow75<9l
zXS!){4rdS%cjx@Hl3`{jf({igjK4nu$65jVnG}RA^hQ6ygw|cxFY1-;Lel5H<{;s~
z^fY=U-wNT4EG&i-N@qQiWSG0tuYb8oa;QkIG}uq~9c*2)WQmNTV)tNjHkZih(}Y-(
z(;0{d6$L-eGwZjVr$0YXLvjIm`D?NI;k+3BfT$}1S{h^Y+J+CC;Vw3VKC0^Q@Gv>;
zBR%Yz1Ig7$0bA<w8e%(<Gb%7p2GXW1<Ki(v3B-8ymnZN_6le7>gkp6hS5=FBza>hA
zd#1#gXylYjyR-93a#9Tn(M56bhxlytJ1w&daYD##Rw||s*PTVBwK$ab@netk$V@BC
zJek2xy!2)<xBQD^7@m>9!Xncc=)O@U>Cc{>z$64c1bL#*{&D*qg-}3}#)rg`><m-(
zDa$P~L`t4qKSGk}iwxW27=QWYmkRAhz}F8~_v>H?LFVBBXe;T`Tm4(Oxwza$n}UAy
z;b!ZWuMcZrGO^>VIjkKB4y71bAU_!{63V5Vx5uHWhow9+($Z(?tCyWDTs6NLv!Emk
zfGZ*=hhVuR!T_*C^P-1z2H1!b&JZ#mz&t)YOln`GA3r;LcwUlnp-qm<m$a3Z(4(I-
z2+v3W3P%(&F5k$bS^;ECXfV^fQ#Nv{G_Qf`OgdU)6u2#MxGJ}0cyJC}??Pd7(f#qH
zuWOPMzd*sr`3LB3O_6c?pJXa}WLu&B8#t+hiB+J=q-P7wis533zS+{+3K~oZ7~B|g
z&Q$dh2EicY$-Yn9V>^+~d;9inU8mAY8J2-HMKK+ny$-n~0s}gxgEYldvBR9Y4!>_9
zhLSw<uS(%V3+J^l3RH4q$#U3l!Zn|)u!I0nNCP%sQWqGNp<aFMAOY)XmP5~E)&x4t
zr9u>I9~dy4c&PGYAM`w~xh8x1^j$R3YbQjzhTr$a-9gpkfLV%Vnq(uS38{$;n!*(q
z@N31$IGMacQd6h0A7Kxo^}uMPUmf!gIm2vqikQ@>{(VKMIEEANsEY_frS{!DnQu~6
z<QBz&mmm+b10aM9P!ge)fCI#6l}Y{<Qfws{m2W2*NFvP3eBgKu4*sM@5WhKcJIpUs
zlY;#GQV2(5$=S2#!3(P)6pZF=G5i3HfSl$MH`&zj<2B7F0-6$P#L2}~4ip$Yg;2Ww
z<lvK1?Tqdu%%^hGNLUQ0N=paJN)Aty+Lga6(DYkISXdb5F-a~4YQx9K0fZ-gj#$J}
zi7w?ijsa9r`jsnJOo8*Fv``aB*+e_x9N6_L>}b;#ziKlxky}(a-Q+|va>N)3f|D&t
z&<s+YIIu*a^-|KXp)4BqLtPgOR4%lIFsM%xL<h>k=@8`TEqswS1ebCrKYudQz^YdP
z!u}Tx^$IIky7jrg-(@U0CgK%2oDm69q>#uspqj&M65{&g2qqFezz9wWJP%?%!<u~t
z90pE`ZC8PDQ#L^J2FP(F5{z##!wN!7R}?nT+CXwu*>ht&<oHxb>104v#iAh;2dH*q
z7Ln+vFoP=%wWcR0bsz|lX9;<z0)9O?3yd6Gg_YUX-X4l}jb-k0#%OG;u{i@EUXG4F
z6-g+=6snJW%uqx`#M)_c6c!L(Rmxo?(_%0Gc!}TUf6~xQ9Oy*O#sZ2X`E^t<E<~K+
zHI$Q@4=IhH7CA;~GUtWn1*GGNbKsz>;*3{o(yO|gvga8C2QWfSPGN&VCAQ_@HR83P
z4UCnXWat8DelhLaNK_EWme>WMK#?5qAn+Q<fnfo*!!)>s(XuuX#S_aqO;wg<r0lDA
zI3rCs9==|iMTCck=M|6kHz;W)q!#Rht^@Zep>E?3=34h4PPYgb%F4<rpsa^MPr2}^
zVKd+HDy)Gc=sj<_j?SAPtC5Z!8w|Tb*{ByY>{<!2nFk{}z9XmZzlKm!9xsR~l>dPl
z@!c~Bk13JkuWq_x_^zr+@yEkG#`iyBreS7v1L{m^Kjup)LC62`n|4`#Njxf0pVnBF
zmfa}3z>-~mYAZo2Ctf*&&Qe})k6lOr{Y%Hh;-`2k{pIt4;)A9uD$VP*B#<9F_Tb-=
z?gjKm`In!*pvh7GZ+?uU2#t0RYY&FkmNfUneAc7d2&zNm;WOpffaMD!Nqh&HX?W5a
zdp9Abqo}dP;M{^U@Zt(DlZ>irV~NIRc6Rm>$czr>JBaJR<WhY#RxkJJelQwnk}B~H
zrSkQbf7ds(_Ip=g)MCx%{UJ}qA`W0&074IFk2bj7uCnl_3qh?v$k8+~9!{O~TC^+Y
zpJ)atY~*C%EAM`7u<nT=r}ZG&hm?LCTG=66=rZJ#BNAjHm^n8$w+mZ5nQ4S16jA!=
z<e;Gsn094j%hUbm4o0Mz*C2rA=<6j6;YcRvHl!sn<J1+#_3JC9k75ePaAK!kC9;?M
zNxr`+tHHnlu~9rwg6sYzC7V&KyfL+4usKG7z8%8peq>K#JODC*EM7{Zbq_ZNW69yE
z2a7-9oIApMK=<*JL-65b$h^BDJsB<n$Hx@dxwAMA!i!3*0MdrlU4=BeE{Kbh!$_d$
zli>mK6YxA>o|EZ6X=UZ!7yTGi;lkOBIC4W05$&t{y$M_*F)q}Dh-y34KBsaZ+(_H3
zjEwq@9An|RZ^4*BC4qWOdz`=kE)Y@~9LM#)|NacK8;X@Tr%o;idQtR$?*+tLYeMD}
z#gV>1TV$F8w1waxq=`tp24<9pw>Jk3>RTCD7nO#`V?EUs#?7Mi5g}GkDl(se6Eiq*
zV^CNiDprB&%E-$TgWD-d7-!j$VR~}LksZEo>WdQ#mM|v`9!WXcUAKPV8q9{OmVgBk
z{T~dRWGr#=VjvD--9aOSrs7i4Q|pC=dtC-Wd(svs8f`G1&Ma$32*p(syEp&nDTHPs
zuVg(xJHbkW51iUk3kvY#U?8~X$%*M{X}ZI0`9#@4lijMyzZV|PZ1MbzD`Z6~4VnlE
z;Gj^my5)!94b5|oE65!*Y}`$={`%A@x;FUPGvX%^E{uA-i1Vb`f=t&OS#Q<pPo_c;
zzSpfZ0O``NJ4kz(cCZ<(p3DeftjBb^FS+j1#=jhfumZF5Y;FXRgA_#~!)5+nI$)*&
zhd4IO1*$zBG=W9_*mP30PIOIDaoYOLn>o6MZ7{F>vU|&qvvo99WDoTq9sG08I*vqz
zhnE0Ck)I8c%+AAe0TT%3!p5>zeK&$7Wa)5xVpBRG`DmaD;4luS?N^cMrs@Enn@{&G
zjic`l(|526GdS6suxIwa#qn6~2es`|_W$MZdHgrHy9XBhD<(Bb$?Y_hwd$*Oa&qsp
zLsV-la9~RM?8BRvaclsNLx+u4SXNasouN1$t2l`DjmIL56$L+#oX-LarV_`Guz#PM
z9j(=itF(v+U-b72ASV|E1sy{ZE<xPVuu^Wb@thyjEUFH%FFHTv=Su?ilJKvSI_HJ#
ze#8ef6KV`x(spCy^2dK8Kwe2g;tv{t$;%_}q56=a0^jkLmMh-p*ucPWaT3!RvdHx}
zzlZ}2BsqCtyib}5_zoPH{5K*3SvTKU9w~U|!QI>+IZ_HGoDU2T_W*bH^X-7LcD?oX
zWCEG66Ho;*z3)5s?AZ>Yq9M)@X{?VF8abgDwgfnWSHnJ!iMsKw^7jVwP|DPk?QkA8
zeNWR#F23HQQBnJw_i%9~bRB_z2ok4x(FE0m7e{dWT4686VRwl}9^Q01XXj4=_gKt7
zfv_>yvNU?Bc4D`$kj*P9iPV$hJVG-wGL+%Q(L*qIS*Do@ij&{n#C7rm=Axr0<<}1O
z5Wx(-XMfhT0kC;Zi~}YD2s)Kf0s|oVdxQ}`zOU-G2~3i-QtJmAwTN1mixY!OfVX2}
zeIHQ)+)-A)k^}TG)GgBltBo8-O_M^vF_vZ*zU8leXN2$@-a$-3V1etOWWa$Q0v=1S
zB~QlSO@iQj4bo^(P~Gv$$w`aoG7%{$^7M*qv__Bv1qFrS@o`x!bw1Ot5+t;bx<$Y<
zfcNgP6f8faFm1+rO9SU;w8+E*Jp@Re{8tHxISWzGv{-0D8(K_SyQk79ywDFo#t5#N
z#~H#_xra~^;p3!nAR9-{Pr{+PTv*s-ULR3#&!MsS;;7$XRzpvP@Od6CQ2GC(?#<(B
z&inuW%or|=F^px#I!jq1W~nH`Y&g;&l6GWCi!F(g7&BvL5DsdTbxO37Htm+dSPH2V
zt=3RlrX(W$?$0{M%yoUP>vLVd&*%2}<2!$R7IHf0ectca^Z9&iPmb2d*B!Tr;XwYy
z-Zc#iBmUqRYRJoYsiP3k2QXV|tf&&{dh<{}?%cZd=EuFy9eG{TfVloLsei@wnBWWh
zfI$@5q!o#O5E+noqfotALKVd`M@rrjWH%yIaRBBHQUOHk0tEtNV319av}rtQP8`FM
zp=4$z7JAru32)#t+{oUGQB8yZ2rAKgN*INt6ai2PQo*H7NEr{xR#BX0!%(vm*^``X
z@T(6n>SptBMpl+YUG`DU5KAGvqwspdb2A2ssRRYBp4G9hxZec6%T*MkpMHp9Qik0u
zz&uoXA{jWap|3f2%DtaEyzu&Cn>q-4ram@8PUN$f!A&3yG8Z{O=R<y1Rwuz`O?TN$
z%RXq#n36>fY9c9Zj*iUu{=R)VN$*7$4Gz2LbN1}ne4l#aB=(u5q!~V}iSM8m?zlGa
z<v!YbHcj1D>$;pi@`{oEKa;5ZCodd6^-DDoRPCZ^ZvT~dM^mck?_V)W;j8_xUg<Wy
zB(2_H`5Rhg?`c@=|KxZ7-)ob)GuTisxV08NFn0|hs%_TCPTV~m?Iy;HqZ3wxM%f~v
zRlKI7(;7T3cq)iRM>~XfYUxeRm4nlR(gOD^?f10KIp@k7TJhbkBz>(Ov9!u${YH!Q
zCjZJot&E<U17}Hlwe;rdm3dkjy0)*({C|<&{3hYoY`GB44AcU>bRY*PcWYJkE`V&9
z+zZIRq;&#LUCxXyL4JH9I=f?=>|hX1Z;8Bj|Go=SXikIez_t==!K#pRx9m=dL>W#4
zB^m^CGZ}E;zyWb=Q_ZC9-IZDDl5n{UMfyLdH<snP-)Z>jRZp0Jvn+m5k+q~|F=<0^
z)lnrSQdSNkfrV=?IW#p`O0!B_I+_iC9$_l5$gw3*hsk2bojyLhhaj?pqX#vbfjwH(
zHyjEkxYQ|v=I}L{`P2eo8w7%8j5kGHiJoppWm>(xKo%l4r}sE=esqKjkxSw$V`iqg
z?aEZERCY<LQ`t!3Ao@&ali|@0nlI!y*@HTABmymoTGNMaN^AynlX1j_OU^Iz@r-p9
zyk|*kFY?H<6R?~qg;thI_>BmAB)BnNmGvL=u$NiJVNE^Bp%Ea+Iuh<%ekJDW*zb;8
z#&YE^U$*2RB?}MKHk|%h(gcJQL3B%h^%#|Z&PsdRraxO)HB=pCpJvVY6Bn0)5EMu#
z2XHx(4uB@-WOhA)vWn!@6<=_1hg1l7>g?<spSKqhNqZsUV_`5379p+xHkpBc74ahR
z-m6nC4<|&PORbD3_jNob7$0T)(_NFO<b@U$vC03eeg4g$<I#<;R15RKmc%-aT%h7f
zCv*|YI5PJX0h3Ra-n;jZMC7SVbFODWXAl`Y{<$121%j|w>KngZ1WSfo9PmR03j-3{
zdk&^WoT$t-Ni`^1VD`_u<D|gXwuX%B*a97DY3=6b`yM5wj)K|{GcxqkUq(UuX<Lon
z{e(11)vZ*4r(a6EZqk!KdsrHp>wVsN_uYXCa&3HfHd>ko`KYGzN(qZgqIre~4Hz(<
zE|}`>{`A0zzvy2MkGfR`n8d_jVPrUP;DqdZ?@v<}ZI9jA(;$3}t3~dpoj!CS#|o+j
zzWefE{-apT4<ioXMx$XF2%IS*3?@zrGyv!kDpk^1_uFU3Y92UiB8#_)zJdHgNsqN&
zxA0zFTO?r1Mudi^PlSYMiM4b!EU-(VZ8fj>B5z+r=4h8WcYhc4jzH1NnPhD0EZW>&
zYtlM(iFMm!qYo#>f15Jea<NOZ9O|ZP-&J=z(-%x;c(7oQGixG<rjFKFXcLLphSJ6b
zc<A!z=aZPhXY{k$K6KywMvU%9jz?qT$IB7Iea%FBvGk?2LNqJQ(V+2#MZUSq_vHdH
z@7%WS10v{ymRKgQ?)JvbzxA_fNx?l~A<<jfh)>2r<iwRh_6UUBB5wPD>iIVmc-ytX
z8NQ;C_5Xwa(eu-}6S$K}{MRPN{TmLr*tnp~`Du`zY$?fuo5n(VG~caqfphZZvj0wQ
z(=GRo*BK2aas4<%3ck)6_kICQ6WA7QI?lUzG3etU6&<zcyXfX5>6Uw8A#ZI)ec9PF
zXT%r?)hT*3Y1inyWzyh?vScoTX8RDBRM!PBW2|6k2wtptZ8@?eUtv}4d>O|16V9QO
z{+7Bq^Y`V-hJ?WY%N*9?L!obc3@k9%#KcYzQV~4EaZB$*uYb&_Zd0#Uoy4ru&>LJ7
zi<i(VHfK9A0&tQtqF<rR;5g^K3^&_1+79g6f_InMlhoU*phJ=>?Sn~7TmaIn;e5Pa
zGgZG+r!9k3f&ZpwOP^nL(C~SrU#r*K45SqVR}e_F6n+{W)R{^nc@`KvC0Yukk*ZJ`
zSkX3Xiv~YC>I5l%X}){ZUTMh<7G(9ln3iY(SpRFtMbA6=DgEkkO8K+BfU=|KhFa<^
zmkupb>2N!jxpU|4d_HO7L|$K6lpgq~R`fYm=j?6fnMq%LpC~2?<UDe`@1ju}1<Kg8
z6c7M$IiW%=cnoeJP2yEL&?Dz_+n=c0q&d#&qHEUA>olr*Ee8s>u^h;ZE5bG_AA|Hq
zePG&p_iL8wMC$#;GH_8Ma)mP{+!6)`+7eCDR5YYL4;!CYrGZEHe?AGyjV|OGCh!9y
z*35X8xmfZ=IU15uQid;IwRYP3nsIAA{p?kAEj^RpJh2d6O$$u=KZFdyB6AN%_ZR_r
zDXldfhU7{~3@2iF9fi1-7+dfI?g?m7OHLVCUz~TPo7UPdJpv=eAOur&6rFF;OE{H_
zsC*%P^L*<6*r20+UpZgxvAcRy=;Eh0*HiEXWI{V3<Ws~a2%=|D-(^<_`m|h&LF#|B
z#4ddM+C)UB^6J_TvD*r_FF=VTMr|HRXeB|UdJ0j5ctAZQ4{VVoXlX^8dMx|;G>-7e
z?`<B3a@B<_vt=If`#zFdChAZ&v_+!Bc#u|Kd<_EHriV_;2g|uIphIoX<nNqQntV?;
zBf62_d{e!qnYCka$(h+mJQ4iC<^;&1lANzwR5Myy&F|Hd3a1QVT%O^wng=P@UHw7U
z4vyHqHV;pT5<D*4px8;@erRd?Lq09PQ%kGKV!%(%bOqW9j`48SQ)>7GB^*=vuf5f;
zaC~U4;>1YiHAOaI-c0ldb<eSWsyylbUXH?kqf>`oP`}vPa`Qc{Q#*}*x}PIjcpsDM
z6R$J6#E-1RfGD}F#CK`KBAL|u{*lL2wHQasHlR%VW@F~!HNkSE^NpzBC31-Whf@*9
zwK7B!zacTUGX)EB#fb)_$HJW8zx&lbxkDg3N><lyW{cbVpk3%7XO0ei0`K*Jz@BaJ
zfm|8k9Z^Ll>#AMBd|{i*meokq9wHq?sv=S(1|Up~e&^1OhbPQs{1nw@QnJb_>rM4#
z$%``XxKf7&Qx^P$1^&r$FU`$!{J@gs-PJ#~d-|TXso-SUlBUMEAMpe|zBx}NN|Z%*
zEu08Hns?B+!^+L$f+Ye<`kYI@d;%dCTh;`|NU+B+Y4rqVL||4P-nHemr(e5udIK;z
zD*eYPC5_S6moM!$x>hrZ-l~t~Ky_O#@j7~LudsK=Jx4md2+XlMF%mpQenhF=uU|jV
z=7^VE0gh&ca>fyR0Z(YflXcZMb?&TBcg@xBIZwZ5JVBLorJ)!8K)K3U9W%=D(_Lk$
zXJOO1-6MvC9jk1Z+3%@A*!H#cqHdvW)KLgSEPVzbq}Aqjg!Lf8k&H|vIzUK1_{Uz|
z49?;qRqpWjpK%Xd`8AG0tFyy*{HX=|yj-iyhMP>BT-uVO({Q@q%lAPfHaI%Yne!O^
zn6*jEn_l|^k*)7j7zcgwNz8?%^vK(^A96LLw;by&_M^QkZ?{9XYv9yptPhWfC$=V~
z(6b@dclf$=6_V~%rg;c*rhqrpk+1DcjYNhNcX}z_9!}wD0P5v2##%e&7DoE7qoBJz
ztXj!UuE^#~`|}k}LBVPprH9@ECKB)kT;!py?|K`hgn*&Oz#2Kdva%Ar)fEV2;TTiy
zp!pR9LtI_VTqH4xw1&pq+4BX0-#%;R>iri=!eW($6SiwffkMt<!7GXi5j@#LTCjaT
zZQi^S-j>Nk?kd^Q<=^QPSNX;$JY6DHV*n5+TGZXetM2#N84xhiw(Zk{#_d+3<j@gG
zL2gfnXJf{rn|O6Z<a|M!1`G>gF)F2z9ETm+XjzY5SS61UaG;2SICzC%5Qmv;(xbm<
zK*k(-mR?D`7G&JBX6XkSPY+}&50QXbAj*-uEy$p^&s?yYRpQk}lnWgCzQBqgrxJ86
zyDDy6`t09w9xo8uA(7a@eJA8A-~2+}RSn)~i1i`CSDk7n;p0G@EU7Svwj@GSY*h@2
z!jYhn(=8b^CG4BmiqR%aP?Gh%(|NNmas4`W&pNeqk6JnXIGH3;4BJMcx7nBX>fTH&
zT(<GF5e;Ncgyk^6Qdx5(xoP-SLg}3(KTA}a9Jo<mTmdNj@4Xe%BVN+2(0PB3#a4oo
zsgd~1`txcM;3T9M-~GJRe+svd{fpZS{L2_E58)3(x1XQ8-BVZ*kp&1<Ekh(8JW^p~
zvgFPf;@P!{SggHbz`-IYK|Wjwx>-vqrMTXZ&H)3=aM?1JCvl6(lV7w4(F$^SH|vi)
zcb}h|&5-{Xx2uivG19&y%`rVz&%*xNQCvC^Pb5+g9Sfb{;hJXzwS|Mr%Bm`QMZF5p
zLNCsyI#CH7KKy%)*#J)S1$$SGQ$8j<sZ5afQFmHGA%%F$4foN=tS%H36l50@=O|Iz
z$`BHRW@;o`8j?kbJF6#+?mG{y6~@iOZXr`CJdSf>G>_o~rL-aN4L139c6J7YUPu@<
zI4cjv5!GmV2KoXpTTyyzi*j=-0&i45AuwSxh514Rk74%%00yB{loDSu7<tW~mjG~F
z0kIyN6ncy3BLC(0N`AL*(LYHLNlK&mE4AH<k}b0DX~oll0A<E8d2;j2{NjGNEs6);
z4f_(;FJkpYkR025{mwwT{t2Qwl>{!TQPkPM%=u!wP)0_t9XM#vRK8H6Ir}2=yD+~a
z0)Frv>v%YTzmB`A(PB0UnB4Yt%k0C^{ntE-IVZ8jXafa_pdc<E@kGMMS$SY>d9bAl
z?e|A}Scm?xd4~>1-tuc^!x{Dy-B8-0)2OJvL2<Y^Jh1nk<fu0%b?P^Ih@t^=g^q&d
zf=fnJ$t{QrL$`dis6mb1oT_skHf~X-Sf4<ZcWmDtoSHfc{Od+zwbi{1*Va5QC^5FZ
z!{MRweOcS@V#Yp`=S|mBYL6Z+uWPz<<h}3S{#MIV=cAopFV(rWe&V2=roSfb2yNpv
z@O(EvtD!5`<j;2w^u8T9Z&%>>Y3>o1^iIY(9!ydm_;t<Hu1in9F}JjSL*Aur@4o$A
z=lJ_~l1r1z(u()3`(sN`ePNX;z(Y6>%ibQ>tSS>}F{bP-f35op)9Q)lz-_`zz^=J)
zPzm7|e+0WEXZaNB13m@E%Pp#2ttqYf)K?^Mp0REqWGO*rIh1i4|A4Adu{VZ1od01R
zo#92#$a_IW<L*>sEV1AE30<&Ko7BSszxWAeW)_grqHtwGXZ#ZxouFNCwEpE^#mw9C
z;`Shm3o9D*GK~DwQ!ng&E%BJkx0cCPi|4Z7hdsetaf<*S-Me}7{qf_+doOPBB&?k?
zg~cJX`l3{l8id(cM*+cj+0#DE{GjZ&1v(#WY;1h_orpwnV{2QPet5lxNvL?m9!zx;
zm4r_0sWSFszR5qqhn&IyB;ou5>z$iv`f2@)=FE*op6OMUf8>q67<LOF>UW@m)sy)&
z+CMD|9r$)i&dp5s&w^!Zi?(M=S6TTH@Fyx@a=B7uV*wnM8I@2q*L967T#$8z8BH%!
zGG9vbD4C1UOCdoXE+(T!H4jI~X!D*@EFRKpr&Od<5arP40e6eWrvEF|MC>xznb*!j
zvTcosaEZPfy*xJVl*3Lv8xb7~%Zu#rnQbFYR0DphWpQ*m!fGSkU)BEyJ&$!ZaZXDM
z?zeul_hu%W6)-1uc%k+&ryT;sP==rplv9*R^o_vbaB7O?4K>g%y?vOaM8e9^FJV%$
z6i@CB+?64mLl5bGRe$g_PXn~sv2EMb{b>Wt0<{$qLA?A~%{&)xj(ci{j)MAmw9+Qt
z*+Q%bxR{JoMzCGYrgBaT)d~_a(^f}gBr>`!!7QAzZfEBuZIt9>sUz`4XyE9ktif-g
zk3vqIIPsdvfS1LO{BN3>nofgNXW`-?A_{vvZE2g=dQS#mk+2kOPm1JS`g(suBqijG
znvFpqloLF@lNh7m1($hv<U;ef=Y;!I#OFOGL>+3#js(oz;+#au#8f5G{o67!eAhi$
zgxK?kH{bh}bBF#djBi}h6~&rI>0qRiTVC}CM8;Na3Hzq$7jQ#tOrewBlf9EN%JFDO
zka+n|MGPG>B8%J>0~FUpv?B7tNXJFqu1)Ob%=$2gjC6^iqAE-Hi)3y~<{XTWGoT}Q
zJ*$Sx+fUDa?WpOu^+2}aENG>kF4}Uh<vv{!z-=M5B0Dqxs9de;LT)lyW5m}k)T1CT
zZQHgL&nej*!tyQK#JV%_Jcv(G?DB$(F#E{%9chX<%7;`JGnP+7L&kl5h*U@R!p;C`
zxyp|%+v(**BMT?V+TL)UE>SVcMpzV+Bvy<N>WM$KL<iva&Pfpt3$Q|Ml%*dH)E<4Z
zQGD1JtJ;{&8Sef_7TqAul%1>`Z(GG2MF^E_3=yjW4PN^fo1S{dUes}6fKqYPYZtwk
z=Tnz!X+iM6x2yWjn8WS`0}OaUKH`RFfvf8izB6=aqb_?0QZ11U)(%frAZwA&#I&PG
zwyHUS>aU?=b;j9uyLb0rP{{uW&S<snNk_U@E<CS6cqeKMqx>N{<9Xm_$Ih<=sE{SW
zfTefKI=uO24rx^#y7t)GBSm{9PF{CKHVuX>o%VKZ`HI$NY|GfYHqCMrs!8Fwp?P^%
z%N{dLJ3*lep)oK~D<#JCrynS~6O<6Ji#1>GTJLY~brBJ{`0<&2`fpcAKnyib`(aaK
zMw6xTgWIhZ1iX{V0h3H59d~O*%&3+UxnwfBn>1;sh1#WIn&~&1vsb<!Vag9YsP>!{
z(7FmS(UHlcpMAwSBd|h6jCw*{GWlIrg^0sXwCcVupB?J!_22`Rp0Jhj7w@oHo+}j9
z=@1MO?xRn`f^RUluT9IZ)~es6_jbGS8Z7_^bvTEBPM<a}e|TiRemLX-@VbbAv5agQ
zwfJMTd`+v+C}sCeWcT6n6#<foLr+F1?uB5&*d9VzUhu5sNOjq=RsEZ8u5;u=Am3@8
z!kWgAF5ne>vtmiUn$6Q{e|D&j#xJAwWI;gd#kkfk&9Ci~rhWyjH~!RQ9iM#TV4JEv
zGes3Y{qTPL&EX+mwtS68ou4_+Qf$!DT9rJYOR6+O*#|dry}?A2ly9j=R;|hoE6l22
z(tSy@Aq!_$pKD&+j?&Tj@{(8Yc}`;MDh{5oEu$L_PjSae?RH3`!Mk}<J`ku<)OkDl
zVEjG<F;6lan$u3_9-aNALKtAKU4QPt2Z)O;z||ggZh<vU^f4{;I2Q8dz>hNqgzs4X
zH?7J~xh;|Y4pTV8tV=8*=LH*mg`tO0l7&8zjsZM!lEX+9T}G83KSKRJtH*wt?|79$
zH9;u>u)%+nb|(Zx##qLy07q92XxZWwh8#IDDpw<Sd)kj*+4Q1zvI6mxgWnU*JDK>B
zIYs@R&~!*1j?p(y)Otj%8orVvUx%3+4IazFb~EL2u4nHGQj(gNY88B@0fbiWzpit#
z*V}KuuP|Qq>4MJ701y<zhgqh`P9^Pyoa;~iruh>3PT<~5U}cf`SObD(LJp9fX1e2_
zLw4v0SN;z_OlrUcW%nU7_pf`st@Y#@f6{4a28e!9(&0Mx5j<p~QZxnL->IiAtxno}
z%*Mt>mb{EnWOGJP0^)*#-rFqNHwq+Rlyuf&ybu|72r7cV`^R?2458MfTj$v|${){u
zFKU@I7Gy|XArD!%S+#)OsiK<(vl_Kz;tiMCBq~ta1*YO;RjmMq0F$4fhOjLtMieSD
z6c)SFzG%6134m7?bxAcc9JJHD7eDS&qj@-%EO<2IRj|Y?2W^cvTw;|0rHC3(-~d~`
zu5=U2+>94Mnyv6V>LhW3cki+|e3SC6gTF<>Tya|<v2;{y_;${ekS-lLqXKTY4vn6v
zt)MqLliauEhH9TR_tQaG#GBbLEX71Pes)<P9jZi3)+f6rdFo`1M7KF_bDceV5wyu3
zB=YP=^z)f(&ek_vZvV!#kk*p$x3Tf{slD6Xp*1WQ#2R`B=%&BqykOHoX!#w6gmrm0
z$a|K3CZnFCzG(&t&bGXbDP4N^?k&O&u9M__t@1MkIfps*7*)9mxQ^#SlHd@6oR0P;
z4{|ayt0n1-cJ(KVmRb(A8vG+32%6R&tM*VUrO1nH&n6*Klr7LAjXe6Yr&L^a-A0`B
z0s>jxG@BT#ABr4X-oe0j^tiIz3wHJK>UUCQZGyIf=Ub5QrHQA$>^zD$jw;$$v0>dD
z^N`k6xO-|Ijieq*91&|sgbk*`%ii4Gd-ez`%V;a1$j$wUoD5u<K@&$PLu3Hu{fQ1n
z>?#U#?oG)%mSLR2U=E$oAx_0}*}O1uTA2A8z7aovrjE{6-lPTRERrrwx(TYWt6va;
zYm%5{1M#+-Z%iD!jZ4C+-AqfQFWj!;rKTGd1X|*zaoF5_@L(Vq1--NlwIl|E9Y7b4
z6nQ-vADUbO|B;OuTn_7ZB9wAx;Ru+h#06k?^V7Gb@?7xO8%*f}Fbo>zeNO!?nnM@s
zC_u(yZ}@V%3V(z)-mq4iDjd6e_wL649tFF7F*0%|3zBJ+4+|q}m$;bUY#CMaBkhJP
zVY&OBB(L+I*u{9C;(`{Yoo%t{mlM&~$n4BRD#53cdO`f~sB?=MauK>)V!xI?M35a>
zWlBolUibDjL+EgKlBKG>kTX((ka(N!QIra!&ApAxkum7nCCAau9Rf$VG3vWrpg*5D
zo*hUQDByrB=EL<RLL>SvRjH1PgHIj2`R<w_ov)HU#mBk#bjEXxFa~VEmLynO5Q&eK
z0D8VJq`qYwxT=+0ASPHHP{F!KH>Q#ApH=Jg<;iY?P6h>tXcnoi0ANtjHGO&^e#I0u
zK?$*L6VmHVy^Pg}!M=!2XDpbB$X-}QL{5uRn%66#XJrvamyIpQ%YJRy8mJE1$b3oj
zXR73kUl2P6Dww205$I>0UIcr~I?Wk=%UW^EO+yJRo-UYe7}^Pf!N4REQ*^@C1=N1L
z<iA6q(kO{6SK33?EK2AfZ`|#yA=_u=!KCm=COeFjNZl-s*cX*1UV2W;LX^DpIhbd~
zAIX>_9g4(j5T}<^+V!y&{kT|2CMYii3_Bt@D=>?fV^p|hBZ+K%$Sgb!Z4^$+=%(ix
zP1gv{mG!026hqEEr7;1B0t@+4UK{}iz@QE_wRF{aYeKlw1CGf5le|T`aCkbg*h%oL
zi2efg#dJnJ2$pC7S{eb*s9pB}rAaY43|38@OwLT3vV;;g6wjAZ?1B6LNK`F7f8^2W
ziWd;h$`JbF{6~v<xI%(_?UMUk+>pUb3A~$*i&0$J4Y-bEk?Z|cuO*Jq96Aa{TK>zA
zaooU|5)e^hzV`rlf#Lm;BSwhN3#O3Vs9Aoa5n0lhOrQdaYfigp(wuSQ$4B6670{5z
zUz!gw2Qwn&Qust8Tiz4kbQ(yfTu<gA7djXj>qKvc>O(FCQLu)N-q>5~F!D&se{+jO
zCn{KwQoMUauv?!a1Wq!7Ih%8#9gyhl2eWv<uaWBrih!H3v3GfoQY%nXF&9lxQfa$l
zOcUo_UNQtWLrQ-hV>yN!=x&kf7)!W26q2BcOO`B&z8j&9yzK(IOYR4e{Xo<#tZf`9
z0Uw!rMq0{d93pmcHvh`}Sk6~WR7bG*St~>y8XxBZKW}&Y*~l2vZ#_9wNZfPt&+xjt
zZG2>t9pl`{;=E<OUvk1F+;BMX#TyFUD-W$MZdl1XpgB-s>psY_BPguo^bn)#w%2f|
z-@A}Zk(^VdUg9rem`rnf)N<6FIb$9=o59y8e+wxza?pc_lsprE>hjw1rbO97#Efu|
zh4#&hy(JG>ArY4DcXu*|33@rPq2%Njik@`z+6%F_NC0Gfk(0hG7ZPC%9|z@@`x!9#
zgWy~HRM}mSgz+myR|$nD>z>Q6Lz{@3Qdlhz3qKFe&sjcW7Fk?haQ-saNU9J=tHh$q
z=)YfDN?}YOr5}S}lm>b?JYLwyiUBl9N7FAq!Gb$%W=V&KrI&BQC&xW0YH;&pR29{d
za>qy8{6^9UAJLipoDjF}PN@eI6kM=*`o*gzU=7?6GHTdYv8%Q8?$o#=j8~X*Z6&kD
z2W(t;cgfbFb(Vd4WbrvgZ%;kb`80QZ{H&AT+8%+drT3^!h-qo&x(0wjvsJ^ya@P+u
zUe@Nt_ud<VzktkSWyGY@oUS?wjv#++yEHEgpc5WHe+(2?ZBdLP8^ox{ej;rJ!x`z9
zTQSn%p7On3Sg=fY1xH;{gX>Q3t*!$?dg(jKwmva4{r-uHn-3Ahv9pd-uP;d20~URO
zI98sh_!}jVeu`|yc=x(G{7URi9C~H+*R#v(Pw*Bmk48eGxV<-gjtp-&(*`I+rQhN4
zmpx6q;<Sqr>=#e%R@wus)YuD+6$sw|8Y8d<EF!YhLKcUxtvU>mD5-Df7Jbrki|x=y
z8OQzOM7<JYRKjEcFX|LE&pf@hPw(EMPGkNq190BDJXYnx<H6s_4sC8NMOLaZUdf`!
z`DQchlb6etpk~IbIvKS5vu(NxhL;Pu3kLt8+gR1>p>I@J5WN?mf=G#B6cD>9DV03V
z%c8gz=O@!fNPrVN{Ad8B*m=pS&0}_5;?SNZ76p$OVPsS^wcpQKPd>T!q42xXxXX}A
z(aq%xciVf(AShF!yCTn>qZ(L-_!JcuF6WRgI==LD={9#$pOVxI0}8`A>18!#=4ftK
zPPc_LY7Ni_L3h@h_x^-kfKy?GvvOZjC$7I}+p@#x#9q-AicJ-Fi3lFW@WFXEm?)UB
zwWm*?7VKWnE_wP#<2GHPbq;}-azd1-y6eiv3OP04_1fo##}8J~xhFN$Ol;rl%{B~V
z0tFS||LjI)2Gs?Bt)#JvPe(*A$eOK8msuA+jJ-a`XvhQBanjQbD~Vx7O3ME^cF5~(
zdsLZGE@aoCUGnm)a=uW~T*xhy^Mz;v+4ITY2-%)ZkjE^&qeUDvtDC~7zKvcw7p8AH
zeN1l;A3a&}s{kRh-S%YLOH0#wRfcUh%QyK=hGVvh>W1RJv5R5zJu~)zG<5Vufyj^;
z31mT`!kjYDL7zyp2J~-|Y}}1lQ1-k`XmD8AN}=-JFvN4@4Z&|OkadeeT^59COXeaO
zZ-(zA_#fM!P%}h;DvH9g_<du1?fY*ZbnaFPa%$)Vg)9LPQL|&J&MxnWQ=?7S@Ez(P
zeMbUnM?bu(e`r;ki|t$sZ62Bb`ngh~Z3FDR6xo@bn)A=-N+dj3mY49=86-UyNRT1@
z>*83xf90YzW8K8m8WNP)x%3<l=!2`lD&y3InL}qjFL!YId9b=G_WmB(ClEXDf2dFt
zJmlL+gp<p5^BEOZ7%fSWpneHaWK^TeRZ`=>?tKn#4?%$~b0O+>l>v;y5e`;Cs157m
zbB?=R{q^hUbkNHiyJKmP6z0?youW(ch?<Is`6OjZcBXk}Uk=@*JqR?u=9gYY;ISyS
z30!0vQUi`(i~V&qa}<4TYU~q@@j0a@eY&G}iq1;Yow^4l?hO|88ijGO`y>!qsjwU|
z-7hX(deLD9OzCr5ADU3Xs>#{dJ9v<bDcd5!ffne6Ra*+28De`TexqhW@cv%s)_YW6
zF)$e$Cr_#U=)(`Si!QZ=<3X^do%ahDK%ul>dF%V$C3CXKd<K2dUU*_{E7cR!x#Oko
zZ|Hq~JS_+6Q)h-dTBU9A<0_(L!8Pxz)Tw|ujv7fiuWOoX=B$qo^9_o@928{PrZIoP
zo4@wzz7fd%o^x!fB?3~JY1|)^#ygna2B*ZpWobnS15A^x`af1}KM&j}`U&?PQ76()
zFKTDM5wWW?3>oT9q0ohpg?f?^TYMN$#2ek3-O!R+%O6G^T7NL8Um0Fdtn`Dhv*t~G
zg0<{RLIO6_y*@WA#>Fs5;81Q%lXKQkbXTbw@+jYHXnXq`bxU?UKpr88Z^`Tk3Hs^c
zsVtsUK?so+(bDa7nFPKpei%Lm{8Y1LofG;VT|Oh0xNP~L34Y8`{~kqNc5gPLqHGDr
zde^KmNnSn`Vovt;kx=eL|7}z``Phyfe<Q;WYW;^!{i2D}IgCh4EXd4Y5zxTS+_WTW
zyy#=H@v74fo?ez)1&hF#|E(JRAE5Do)a+LOf!Yi#lgCAuomsM2l**e7|8t117b#@u
zg!0p)0Sog=#~qw_qcJBOibApwXPwXzAtXqxjAJ59m4~Tq+k#QbF*L7=Y+<?9uGLbQ
z<4<_2@2JC47(^?>z;XA{qevju2Z&;H9nw?T2fyC7X>8%bH?`Jg9J;xzVxqACk^9GT
z@da@dT|!>o@XHq;)&`X%E_R?Ad?P!dILM7i-d!qesH`6bAd@v#+&kCyf4&|kN099u
zK&d*6El#pOiLgX*@UUJ=hChix07*Sy>pqcmB1T9T1g3$XIQ~Fv#(_`gU6AGK5^=`f
zlpT(%c~s>AMXqZ}(-J@~Gx~vG{*|=jT&bO))UsVsQe?P~JRc7cg#ynQ({mKu75exK
z`FSu8lKd-gCT7m}LJ5E@3MIh&N6wD-fi-;-Mi?7Aa}G-vA(nWFfkVl6Ww(*L#AdjF
zR7+yiNky^urlNOJ;RX#C>t5Oz)yZx)bjd<MJL%7H9R4bHROB2~c5a)6>n3NBUo1VH
z7m6E+sY2bjI2nu=Z#OgCFnsSHUUn$G{c|g=>YD|3<Eo-7^Q>BVdW_+#DD>r_=^uN;
zHFh!5B=Os4rn5!@VE=wf!ojZ=P-klT^q^?cEq8lAKL}L9fP3?`&GEa(`Y-a|b9Bev
z<9h&Wcu2D5JAj&_df9V7_1n>+qz7~OWkC}WBjFJ6GhI5|KSXo3>SgcKr%u&8@&*d8
zePPtLO`FmNLvLn=y)LwqcW(qZtR5&w>9y=^>4+$YJ)G^AzL;faYb$6Ub9U_5JC;$_
zI~VMWa<KLOt);uFeqSQ?;p=f}UAFb?+3L7E^_p(Rf)XN>#v=`AU-_Ag$!JoD4Vb5&
z-)KGX=Hk;s+ZBUb+H6mu!g>88@!>s_N>qp97Ya-vJ4imBHCC~q^{6S2ls6@YSx2F6
zi(Vo7m9JCmWkXU<xW!=Be+SPT`ugkDV^$b-@&&z8nDbS?A`vRzkq(}_{Q&2UEJV@i
z^H%i<#-6;IfOo08-a|UId_=9Sd%C1b*p&7l1B2YY7h5`OEv>nST2~<n(ou+y!Ngw!
z(f<BTyJDVFo*7fBr?@kSps9#aN7ZS)h}wDV+^=#()3T607a{?Cd{ySi^g)UBB){fa
zmas?_ommOp{NJc0d}2EeQhT2|i}EI-h8O<6^@BG8FD`UY|HQy|foxkai@sggOrAdI
z;oDm8PhJsrvKNNX*c+?IYec1EPl7dT@a&^wx)80~{NIv&zqV=lnX#i)llnMhf6#Si
z!GlCOH#~RQbs>$!LN?KNFj>rDDgp3)K2*cpYuj$&@0qYEFS?IY*|>cD0}U&&^5Yvi
zCsPP_+BGS3)bqt$Yre6C;+BIBlb8{j0STV}2h5shttP=yfDDluQqL`Ue8oc}cC5et
z>{(@DA^vb^f6xd?>_Q-ArM#v7OecH}e)UdjcM&>jL-B=u_mKuPetynm+f4@5NbQHD
znd#k{p#B#(UoM@jy>ggC`Va-_C3D)1*FgF8Tf9F!`J6303gwH&M$-|RZ|1n~4!%wN
z90BL&CRZF6+;$ARi%^SwplFVDQnMGVi`vR}h^ZqPQ*nEPbLeV?Vje%DQ|hbZS>A-s
z2_A`40#3wndLR?pxqM*#L9=h3SC*~W(`|`mz54E(6=dgJ9ZGg3b^i@7=@CUM!`)xe
zxqZurX@83z4}gw=W4mbMUdw!sBrlIF&YefSZxtmXU2gm2#*(Jki|y>e6ugJ1?~0M1
z?V6;@4iW%L_&gJn+TM#P_+0(+-QN~oT%Hk=I-t|-*jq1!oLhAsh=Iv?IapOnQcF09
zR(n~$p=5z=G87T9*g86AjQgG_npfYtikm$b0|sQVFkVXB1+2GXo3Mc#*}!-So_}k3
z_X9)d(qtzP2=!&pLQ$qD%w<p6xS3inPc3R!OwS}PBtlfdg-J>G@_Q)~DD0z&lC4h5
zl|NW5EiG)t<1F*0NMIO8YkCidpJ4JTLvow4@oKB1M{T|xPAdh`ux)v4|0?h=N%#jQ
zZ(gTYyym$ZYM}{A@d{9{^6U?^yvdtyeYatQ>4lU@rvZXIfMa1U7)I}wH4*=Yw_3J?
z6_maZ1{}J9GJWA{I;JG{HXLSunSd&S4Y5~1=kkbmNOoq+DQyNO=Wtp}=1+fgCGMN;
z=$6<j@{vNc`$C~h<_*_;1~r1sz<-RYkC{g?Eh|n-p-+z_gy14Y3cbJu_dQfr7k&f=
zdM`Cuk5?D-BVcmgb*w?olpf^kS3w>n|47!fAd3_`Q!|+)am<}MLGcFwj*G^3b%ue%
z^<S1Qjs+W{!L$bO)0XAfDt=seVTou5c+Zwp8KTD?G9-x?OPlrHQ-$$1Io0w2p6GL!
z69o8Ng4%)_rTlACF0$tPVkL#a5N8&0S=+`IVAWrEV>jBxK!F#5Jn3C5L2kvqAvh+?
zR;jfkr-}?BRCQU61xzaPV@QZ?=*ht9_Bp<gBm`zQN;AN)wP18Yh9Ku-#Qg?9HJ=cQ
zb;bLb@Fa<dY=CX(qk;WJW-M$LKf+fq*6$Cbfs%kIk_wD5V(jBa#Y7vxs6^(H`zDEY
zV6+f*CF&1JJ~oHS6zP{hp#T`ttneU){`T8fvdoF%CKprC3yzQ;M)M1wb6PNfg=)(k
z?m*-m4X&G?JNsK?pQmu87i#;dpV&g5KjF{JPdob8ya@ywXH>#d*b)YFjfhtF$4CeS
zK&8o3O`B|@^RV=%5|}#yxXERdtyhvT$rTUB>?FW8VzK#lj+9`rvH(zdJguHB*OYgd
z#~mYKB(z?GhYYzi-&L$L;GUwwWMZ@D!)EjGfY(HJ4xsu;Nqv8o^AX1&wWB3=s&$lA
z>^Poz6?d#SFZe$0|MD3*Uue-6Vl$IFlZQZfCsFSy#m#@Ca4=pX5QJPXQHk>9nLkW(
zBXg<oibsd(maLx3hHpBOfhHy<mZzS6Gv{pDWe!!Um#I=9)`H^wPWnCN?JKhB$}2QS
zbmi8AlTI^7yT|n#v&L4kp_eaTo?fe0Qv4E0$+EaAQjPdvQmBRYQHb{I_l;e?(r5v+
zC65R*#4$e)GK~1nbrq_`n@zLn?R5k`o5Ej5Rq$0&tAs@3PI2=e!y&!F-nY@iWnnw8
z8YbL@^s+Lkfz8$WME?UYnU3~mf_b#Et^f9uENnOTdm?-QA%(oA;bh|sfJc>q-sO4}
zoz~A8P2X_Y#};J?D^ZvuW8dc3d@bxv8@GreY4Z`*9-{5CY&p)BR2`mn$Pa$T-KQ2X
zKEr>9=c+av5p;<lMEBY(#eR0}*eGR-hiro38S>(Oa*vu8j6Dy~yg5fh6p)xm#Gg)6
zJ=?uX1|TK4dZ?|>DxFZhb`KFa-Ft2;9XPl0<KNzyeAkJ9G4en5MSfC@1Y@RB!h0cq
z?mj5zOoQriQ+=9H*O%1H`!^OO)t4PPe|zLl@70f{<<V<b01VaD=>-q9jsnCdUPJL8
z|59Xb4Uew^jdEMgciYGLffyIm@8IfYUI0Nd4{|P3)CUTcS0d9}OzE~Ehs<gx61dIy
zCkLiN)&T>}bMQERHb}Q<QhU>EdYEN(FfP}3I~p0iK$L&?xvl^G2}q3yVD3h^BUcP%
z*I;e!-%6i~id8)M2+u5%md!(!B^e$(?4L(pOGu<VCkLnB{;}o5ct3vd!9!o)^GSPv
zGRI|CG~I3~_G?ovE6eGWHfuk`Fd_#jX$0-{OUz4WgAFS~pkEY%l~A^#&flOgD!!jQ
zv*+ejhZ2YQkY_uv+6%;+;KiPk?Lp%35+4#kfT>nC4NFZ!)+%SU;r-+61fgl=N%{qL
z;E>q6xM?NpMKmJ_wY!eU8o6}9CwGcPZNkwYoIH$@H*t=W(;cZ{vI<a<&CRoDi&_&h
zrL}-ze0jhgx(|Kp2O{-LA6v^QG7PVp%dQ^;LsN2{=k%F|LP;U}PHS!N=Q;T1LeJ_H
zP4b_Z4e#fwirOjLQ58yd*+-`X@&>sNwLOk4WqCn%@L%+iA`FyNAv`z|%N@>;|11m4
zEgW56JSj3lL9l9U2IW{EPSYi<wfDe@myUbVsg@HiM~s<8H{Iz|sGkzv&uut!j4MRG
z98~0)NsEq45m>|=fZ>?tZX5U-AhTcvltrPExWHnGhT~5<QA;Asa>yV##H66%>6;F9
zQvNQ02zaOqxT4X-&u8-C2x*#tS~)iQ24k%(&bIpIr1_CV9+)Iaxp&R7B0HwJ;%3#Z
zB2k1S!1rn0L=kG)1ft}c04{Ugt7rZxuq4DO0u*bXw?6yp@gl{g<72uGI%&hE9Gb5i
ztA%5PJ4b@sJdx8km>--d3rbPSgwKrmwRq*7j^#`1<Kpy5l}Mb~SKU;eiTI2D^5e^<
z7izk6?)-s5I3%8u%w?_la7t~7AWdHV>{z@B-~lB_8G?0iW`UmKD#zCY?qQYWUH%0X
z%e;16yf6LT^+0+LQ_o=xJ}6ln2!>2^Xn`=)7<-LhItM=~zLS))jK$5DVPGc%l=y$3
zE{sZT)4X!MjE%X9mwh(g=G<>QFri&B9|1we{bzJe)_#A0-=i5yv^qVPUK!uw{Hf}m
zWbK}uzXDc@>;IR4m8<<JRf^|g37hpo6E0pTC|<kEqiQ#C8Ci|zUZ8x#|I>G%DP*Ms
z4wqa~G<FV}HVzua0`xl`qrS*&`78mzNAOV}DCiz`0jf(Zu}OWQyR?^axKBw^FG<k<
zYHC-*`Gx%^cAHm2OGdf<d9d11RT%ttppd0VEg!#-ds&S4$T?v!G8q;&PlN2xhlmom
z7ITCAI<W@RHQepl52d5{3Et^H=?W!OfE`%@2kqb``c#7MZWJz{5tSKV%&(H|E-*S_
zfT57{V0QNtE?-M<vq4h!fQdhsu|^tbUR)NqOVhxFevpwuOu>Tfkb@w9LhikQKVkuq
zhkNHR2Yc$GC=JC#(6!7&FH&kn`iD9z-EzxY>fDJ&8TVJ&TN@%WAgSc+d<Q^5>P)R&
z0oc7##2c&6N%$^T`i&o-gklNE+&3oU#2q~the-I_q7{WA)vGmO-*T4w8a4b9kp{_4
zLu$Y7b_f2vc0uPZU6czP7?^7xrEHhJ8KtNCEkp_MK$kC%oCK&THoQLS<I0~}F9w1W
zNR4h=`IcI*xZ)_YsOXU45_eo(IXNU`ZQC!lP1NuD{S8+YUz9$JekHes8n8`UkOQ4=
z>iP6%tABx@RC;||51%AvHlcMOk&7!TTAK4EJ(w-TcRh^?FTvO7fdZUHfr8HvI{)@<
z^Iw;0ow*mSz4G#IqhKKw<x8iEZi^^ks^#27zW-a1MzZF4o|N$jG3AGxJ_^JthUb~t
zv$f8gWr&mU1S(bl6!6Zy5z?|Ue#pvgX$+=#tFM4o`>>hsq;k{cvk;lHHIxF`za&N$
z98v){x_h@=RjsEsgO+`p$0C!Arp6<D83#qf4xne$fEAD)F@%aD>oP^XqrI(A{pxE!
z)HVOs^S2*6OUyb5@{VBm<8aOzF<3#~0j>^F(2Wkduwv|6C-3F&M19!&-`JnzA97pM
zuPK**7AB;N_4edWCXc7Yv{wI&*3qcz&eeV`@s{#T{_LVqzuU2+f8KG^;N3S>O?^oi
zVsbi2!PTX|#TKs0Ru8ZB5eZip%WvE|z;SAS3GSBtxplYo72@~G;gMlp)CB<{pFU}*
zy#}1>YPYRRD#8^BPF?ftUXMMaR+%%0P6x+nrgJ5hm5-}S4H9d;j6O?$(^xU5bbcqb
z*^?m8H`@;cBE^^gBcf)V=1yaLKc5iFw9PrRLYud1<fc<vpILD}@OzBhfM3E8h#yik
z84`%jm`Ku~)j6<6u_QL3+Imd5>@*srl}k6Y;4D+-{uGok_~F~_FCAHRQTbM@6LvFP
z2ds2kx@5@|>WXBl%fcCMOIa`!)yUVMO^RqUtj?4qAIao&%T0_f@Mk+d1jEh@a5j;3
z$W#HA%Z!-F0*ed>@NUn+8VTm{s-pPTRZ+~jBDR&RBGH`SNR9%x-tpBy^*6SpM0XoM
z*4O*dC!&)Pr3dO7Q{z`Jvv|kKZ}X)3GDIQ@R{yiY;g!2uYY@2oOD1R*Z=LW7pRQ=D
zRX8XyFj}XN8f%*J>xfqXUHzcH0(4*T<sL8p_F_DjNM8Boy0Nn2q0l2}_miUdG~>ZH
zqbr(kY0q97UGOZA7f+W$G>hDk>u|yI5#zJy(5_s$(z}8yDd-S2*Xi;U)0~o9?=AIx
zU-QccW0_!UJIziQj{+LMap`wtV<U^W-O0__5o`h~+f?t{WclcmfbUZGc0OL9`CunP
zW!)Ul#rJywI!z>xpjE4;!2y(==`1@&%NR4_%ORg6?Cm_Tod&ktYD(uL?W9~-?3LdK
zKF0$U;oJ05Qlq9V*Qu(i-gJD-R*h+Xr~x@W&>cDoC=k{a488TE*7YwmD=!R9$QG0o
z6poceW>32RCUc$bSVK%kwC1Kta6%$K1|XoxHy(|G2eI~4XXkJhvPL765!((`$4F&{
zh6&{l@tY!JqM*py+9YQ!8(Wr$%32VFD$~)(5xOf>92Qoiv!eNtOg?S#-SggZKiV?t
zSLVzJ*321B{<x{`y$-cQz-y-Q8)PKd$zBxPn@p2a4osf-dKXP%#a%}+x-VZYv?7h}
zl>J7&X={hj{q$R1y-qufQ%nF)86C_2sc!3F@1F{U!()e;;9As6jxx!_5l;oQH&oYZ
zzjuWdM{g{|6Eb;JUptw6h;|r8_sWoGb8aZh#u+XJYk`zGDAT=YD=AYv)$EoCX1clm
zeA62~{{Id)Sy$|M;#<vQP;-}56=ZkhkGOy@BnKP(gXqbz_E<b&34j)I9Xa}C#j%d!
zKha_tsiddHm@09Z;zi&%66~y5ohkjuvWgOLHz}FG)_>J%(H(*c=fG}vcyTFVug-q}
zS@c-hY3zsUXJ&1|1K=)n8%KnThBFN=dSZmRnaqK<7;8{4UCkd&Bip>DL!Jbe6zFz<
zA~Ef-tE%_G9zzB!a=<sy+nZQ!r(37K|9SJ~Obg57#Cdq3|G}ItTbwlVC5iO`iRtms
z3YM1W$1rg_fF&87^&|^fIq=%rFFw+|wbh6*_-vDuG@AjVT#CDk*%y81!`^QHpzhhd
zz^G3u2d0Y+ih1pYkapC*^Pi;C`O}Oh_Ftwax~}LF7%s(dQ=_i;thjey$$zdAYgvbO
zUja(d=|qL*c|GCAaBfD@bcQ%IxPvYX^%e`SWdUyooM<}8ifucqjlg9SluQY8-Q@Al
zr7VeRjgG`0mvypdl>lgnNpnk$LE{Vy&_8P{GG5e}Zrl5{67#M!p3D`x(8mCTa{4ou
zughz*=xfc%@W!Wn0ra>EjwqS^r#>n#vG%`pWYgKyYcu`Rp)Y0@fJ{oofbN?_p_wMp
zcyJh^G-WcpLQ+Tj-gMnt`R))!uDgHT%O4m}oLPdL5$?qn(QqI)7oKuspv=Gj{kI=H
zILv3BsI+Iz9_K@5rFU{tl0wNmzCB`o^w-$@v{^ki+|mI)z8haEp4$iHgUZ}5_$T@o
zn~NXL#J*Ty+r*zU!lC*Yav@l@`eBUD26UFtRcF>oi0gbW0n$tkp%Yy0RBhBee(fW9
ztI`_?$1MToD@rcY^I!`^fAD+kSErg~U+CvhL`8l{^PkH|-TMv}BcSByW_iQ1Jzw~3
z+!&kmlJ}saU<j+}GiCz~DPm^{BA}HiEz3Zf)aS_NjNzIxHgE^e`~^}sC@Ff63r8B&
zxhQ^L!Oe<oGcEz|x+tfB1}s@m6*<6_po7^k2oC{SrlW=Qi^psd4yia%iH2?2V-~QT
z3%nG6od{%hxD6DA9pRea`xQwA{)?94G3@>~h%}WHw8)Xgulr6uIMG<^<~LCd?)U8q
zBHSCtSLL_tJ1Q&({t^{u*Rro7{!RrD{`vYg98|hj3K5LBT-Y%PS@Jd#fjG8g{j3By
za)7FBZIJ;-QFQwdE+9%LQEkFX<DQ!W(t4FEBM<I_2M=s@dWh;;v@U`P79NiQ|C^`R
zgrmFR!+@z+y?%fM>-zTF`$gAKP+*hM_#{XI3?$AUv?o8w{ScQsR1(+;@0jrXR@FL9
z=+3}6{>@<~jPC<unS4<r3*sYv-homyucBfVY@SCg1kf~$t0kB(St>(@4Lj2N4%SDR
zt)!|6w4~HAmC}($$N@&LM1POtI}Z+BOsNxTlprd?{fnp+P_Ub>y^U~t=-`0Hok5Nz
zA4uGW$i78sAlp~@Q`r#70MKLAm=`8XG>_*DadYhr*i}S~-(6kbS)IFgpYOhzR-V*}
zPb|H896Q7<Kng^O@-0}}9RV%T(Y5VI*(+a-X14a1SEejzp3EAt1+alF`R2*jZfkCR
zt-?f2pQ3fb=jT8F_>kt0Z!UJjq1XTfD$`E(0j=-PYUZ}5HZFgTFrMFOF1FT^eVQ7e
zrS*4B+rR%~Zk|3Zed38Ub>FL1^P9R0Sj#5fwDA^p_n-IkbvxnnRqOA&9-A_y^`Pl-
zLBHOdYHYCbH=E+D!y%`S#09qdW=83}%`Q1R0*CfD3T~(R?KgdAuPYZW&i%Z{lnWPs
z{Mc_oh?8>Cpu_vudRLY78~WsUQLnrHaTU*Ix*RcBJ$hf&{pT+qrMauT{<u~9#=p!n
zU!ZPX^>1EEX>u*^(lhgZzyUWlhV|N{NwWR>ZLog{l>9$WOaDs@``UhHWTF6U0Fhl$
z9n>1Gqlo~Hn@Q*b49vVd3YCS84Q*~t8;e%MH4AFB;(lZ~1w0fn<t6;fF(WnEg8%%3
zo#0@RAQl}OCds3KnD@G~dHD2c(;Pn|A7HO+|Ity%%*ML`x9y#^>W|xBH!sMe0VJ(7
z6u7I54U*f=%vn0(B_o9>reK|96%>jP0n;rkELJ$;<Po{Rv16ZFSXvIIGm#ZF=z=<w
zw7Yp|vnJ8X5J|Eu2&C;f%o!`x4dE3suz+VokP~jqe)<{Bj2lNn84G-}5q4CRULvQE
zDX)%n2`f=Ua`q4~b_o4V<CDaU_+CvUh=?1mS@Q4VA}}ui>O|=*A-zaZUG!NxLAqk_
zLcJy_jUcpp#`qQw-P&f6q2>}EmCQQX{z+08S%wm83DBTBY?!Q(oo;UaD>*skPg(pb
zt2LEf5p+S2ZRliUgHCPDvx^QaSepQoas_bO!qzsD(^8}%@$vDh7|2t6-#ch?<wz4)
z0=CzZ#gc+3f~GJcpDGPB1}Ho>bBNSZ=#b@vkH7p|RhQ~(pQ{u3v?9cQt(11EYuCAM
zvW()}H#+DO)&xx$LGYZSmQ!>ZFfMszM-wE-moh5~&vQem$PuBT<AjF?b&0B=Pm?Rj
z(K(&yM@mqB(S1<%2KpLVo57nm2pf0eJ<Ss`R<<w^q<4V#@5K4@*14PtRpzBjuGxj+
z=rKEPbBHf=g+vdD`<vsxPnKXe&477DgT&r5w+s(6Jh3Y&-a|bGKV3M29Z^8V6s`Lo
zZ`vsUh^dl;Iy)qZ{Fz<`_}0Q$2&Ihdv^d58hO56eP0yw1VWUQ+Dm+?c`MtWWY;DV$
zf~%6#xpRgu%xZY;j*Hb0E)-1}BxnB3>qwDO4E^SvxZF{bIIL~g&O(RP$CJBnJTrTs
zjjipF^GlEC_Y~6<vbeG`XP`cP<9FMvi5{WZ2da#Il+_0*6iH3-zD;X3IL?T+9i%|%
z``eZo=eTq{2L}<57(Mb8{Woc9^ec@{23lF)YToAz`DV1G$cUaczGzO&18!U)3zwJp
z)Po&gs?0}p9zU@o*K((}EEypWC3G0EcA|bieE4Rn-B``tdi=uLR<BEe6YvDvjzsl|
zqg`@<5J1PF1c_%CYf$hzSP{gYM6QNpoFa>J+vr>k8BM8|@1BbzyC`8+rV;oL4<SB(
zX6vM+r40i$VUSwAKyU~}$>RthTd1?R*d*Hm9#!I98PZZ*pkzZu^o3^~o338;Y>BlZ
ziv;o(`a2kk^CJ0K^sCpfQ01n0R&_6V{rG#;E^3$M^>)PS@$PdU`~fPjJ7h?l%N79M
zGjqb~5!<OGT>q`-(qz+bTVHN;`yJr9S8Asvv2*_Sp>(g(cmL-L%m2Ks$IKg@@upjO
z7p?vFkJNy`Uk-oSiT;sVg4A=$^ZIw(?7j$&?{y_D%2aE~`WVgK#=3FMz4|}T+WX&r
ztXUsHc*4!yijY7u{IzIqCiS0Y6Oix3X}j~PQx3W6bKFF<M_QPLmDP<OcDB;fY;n-)
zWnf?s9u*})+Hjk$NXiwA%F@XRbXQhZmM}DNDv!UjiwpUNQEnWS{J7ic@G?>iBKei^
z9`7xAO>@I9ku)U%zB0>Y;6S6bD6MV={Lf@=DVOq3d+9J&3H-(;5SxHycF<x>S-=ph
zZP*K<xsliQe^`)sq-EG<v)$a7nfb?mEiZ>eT+>+Yz>49FDAQKf-89!e&=tuv#;<T)
z`p5&DBU-IVUvyz@Vsf%Ir}JK^ToQzd9W{ruE2yrZc}uo<J(~ZDae>2KweRKQ0i;po
zcxR#vh1PrO<<L|`jDWvd#J4}wZcnVOPqf<}R?&XJ=gxIYEb0&VDNY)R9%pELbaoNU
z2COlvtGB8U#C!y&$9Y#!N1ZCi@+9~-sf7dNc!6^bD>@Ewdc$OHg)@|z*MsCpFK)V)
z9RUY62)+SbAxVc8$R5(l7hh-yG%puv49X@`WjaGEBRMC~^-aJun`Eyc-6}`8XrDzc
zU}3Sd$mCyD{WOnLNrlT>EhfTNaBfJh+?*UqoQ2v{QF25qVKC@!5&kK322xfyaU~#L
zKvT-60l<Er(W|fD@{oK#?zpsJ<D1@XsTw1;gTYTw(yxGrx;99#8&a!<=;*{=KxjJB
zTX}t?DOZM0!`!^0)7TX2qAyFL>*$#U=W21I9hJqRf&mL0BV%Y#j9=elGZp~H2&+X1
zVJJh@c~`^1xI?*-#O+gYcDRR+S(w2;X;PvZ#VjhSV8-Y0goGKGQ}P%F^$8S(M=b|5
zG_2QS#|RZ$)=xrwi8&97%+kzk^XdAYS?8jnIykv;{^ufCs>{It$P-6kRKIn+qApye
z<hrvB+pB1mxch*CF3$S+0kE?rQY&q{XBBb_#*<koJrS{SfnpPLcVk(Zw$SFw+*2MF
z&uQ-Et3R9n=`M*{LW??4DR>%zF%`quK+4c13OLBE-|^Js!<|<xWZ|+%h_M$J|7t9W
z*p$Y@AfNd6WKluL;PK-|q3DLR+9~yi3B5c^G}uB2<xnmpvoL0+xIUdkOpMGAqZ7MV
z3Le!N98}Rxb?jq2UE;)05%2uv7pFakSg%I~kd5c?M>#0K6_z!!nO5G4q!j@&px+-E
zWN;Rt2fAx%m>9VqHQY*F6y7Hl!J{Q@ujF6393O1=%uFO-0~AvDf>-=8f8M<F6khK9
z_#aY~q$x_DC5q|ru&}FG-WGXQ!0t!Md<y<S)cIR#8S5Ai<ln6Lid)4aC>|0zLrd^i
z%q~vuX7lD9<W{aj+LHg>KBYeVg2K7_-Suaiotz3mMGZdQ8AA8;V6tz+c5MR6qo%rx
z3^$`8_ekWf&G4j%Z6pKjBH=vzg(@b~%E2W!5I;m<U{7RtXL?quaz|n-5r-{&fy~!M
z@xuI*X&1@Ud8#Prl3BNE)t+nS^}RwO#FXz@<PiXf8sW>?mba(8U(?eK&6wF+&-u{y
z&50JS`RE)ircDdRg{B|2qWK0{yRRqaHTuWvsTeT8vImNc0v_pDhYqaQJ4`u#2)~oC
zFmdJ(m1Cc>kv|`Z5!n>#jv|)Gh}h50m(yB|*Q5j;n_6q<z5B~OpeqxTmyKJpCp7wq
zfs2`$*-q??X})_#|CdJ5p}y~u(Zh3}wc@vJ)vkMR_4Mo}EThtQNeh@9l|C9kUg8*Y
zS+~5Q5pTwJZLquZp<r&dO2O^~9}{Vs1TNYvO8QzzK7wB_s`rK6?224y@R%_f{T<jR
zg`pyXV?}Jr3;|Pk+#~ipw`MuFtixN9<_JE}%Vb~hO$((*VpQ;F>wBf7tHN_%b*Hma
zX4oBF{pk89Dahoj%Zp#ndKlV%WqCq|=cDUaXZse7nXviB$zB+W7w<bW=PJ2vA1Q>f
zGIJ>MjRVum*~h+)pmeb6!9=vkGLWL~FC!W&7W-b~k*>(cfx`3RlGY;PJ~7#h!2#nJ
zYnUR%ItaroE_9kVs%&7lcOAS7L(k7uF26T3OEslr{;2gQB{CBt&blBlnfD!EX;MvY
z5W==zy?gcU?V5Ks=FQ<pf^=_26<FAmtj{d_A~$(y=2SboC|N;D1|*|P-1APQvRYZ2
z5>PH@4l^iyrECO1<o}JZTg0n3?+~tgdFPM%j3zrQODnE(1P)LSx7FK8>d*&QapBvN
zz(2Q!vQt^1%@zGJE;%>}*5wtuu$)GeW;b?Q*zA+VdhGO#bHPhbvV{ttN2HsX9nVEs
zX{jvglWFsV57uS8y&>1kqfJ$ve(5deIu88?(mO?gN3eepsybQxRB@ILbaKY?qN+IX
zQepb&Mj3;EYQ0K#Em(f8kLO{A4XUM{Rg0kH#AH9dE^0QK&DlNHOw0z(?qhG+L+teV
zb@biY$p@nSP@sxyg~Op756sb&P*VKlxucHd9ZoJms<NR1F^e)pph4&22J2WAY1k3r
z;S(7KmtAi{9Dq!55|emJrS0)mh0|LtE?T7%Fw9}abq5Ozov@c5^)7RNaODFC4{^%}
z0PB&UhKs@})+8efgKu<7WwI$6$D`$+5ioUcnC~FPTU%xwt{u}<J4WJdiy|m0E1LGV
z)Xx{w-+i+UPt=zLel&1gxi(01F8g!L-Q1$MC1aHmbQRm-3H+5Tem?!?e4NY55Q;?^
zjhE^P#}V(vRJjIwj(&J0qcp25u7<kNfXFs3K1ft28<)z4K$~(Zm+P9nvL;ee)~VBm
zJ`moL#`N{qZ``kVd8zr+p3W`%w$R0g0;`-}wf<T(%^-A+t|$j-8Kt=bq$HnO^{8ig
z*6gK^dj%QG?17HNvA$tT)6*?pc`amS^|WeaeN&Z^kOeq^EW!Z^W{zHB;jif8!nq91
znu5i}qsmX0A$$_+h=lUAuuxaQ8ss=ZRAvfUwxa$$Y}>N%fMX3so#;gZSr7o<7Eb^=
zxsrUQ(&RcwJbU0<kqL+ohO<mwKyteQq|wv68g?*T;A0k*3$V-<9TI@9k;WB`?rfsS
z^(FNb37|*|@kr$W?Mw0k<Pvxdn>Hhg@0c!iPY!#n7<Ia7@-dH+Z6xI?$Tq&|aUZZi
zGGd$eeo7FtNGXvn9zI!XHl+JT=1SGgp#{lf44kso-EKkg#(02To3jEJmPFFuMdeED
z9Z=1FLHwf}GVEdBgirGr?jjx#U=|dt5_}9xD~5L<98zCGNo$j34~qs=Z5)<<c=zt!
zqLGKo1zQag>9JUKN$gq?T;}2FnFrrM|1cC3N|YkpSi+ZyCOxRuGbkm!G$)xX1JSgg
zw~#WEu!R*pVCxc*M%9QT!-(^X9n141zYbWs==zrS!+#3W&FY8FRU+d>JiH>sm0Bi1
zE)TJRuC4T!<P=7M`IKRLTfDp@2+$NNwfut(=k#+L+peOi8t(N?OV#W3kS<Lxc{F+j
z;2F>BvZNA)e?dIQ_*8{L0e^9<obdOE`eEVNEk!n3u0pi!QolM>l=lr8@I=lgF7;l4
zvFV{Xx{8)LcKP)@R8NwYhZ;1saul86;;Zgq)z>_yRoui3eh4w#K_SPQUfdrORndLp
z*WgB=HiBgre2?yOY9AA?oKoBEk_eT&Y^8@|?w%N%#OF*Q`Jq2B>rWRu#MrT8H~Ot=
z`Ifb`9`-Ld>GJWrV$*1pHRdQhyLIk-fSzAgiILK6fe{xY$ZXFd=rBD+0~-L3&_f8L
zIePU><zwOkhOfNLLnM|T)jmf@$2t*2$U!e%9vnaz_=a>|NLU9zAn4)jI~lA_9ManV
z$)a#}nnH_^wHnS$QYML6pq?-iaxU{Y8tTqWYxsP=<(Q<aow1tL#?k_zTZ(9US=FQx
zHz6{aDp}JjEeJUXWaOQ?wY{GjJK7(%D4AbCkGmg<rV|!~%{?>Npb)f=5DBL|)%1>g
zI%?!5QMW>kU(MBPzFBmk1r5Gp+u3e>qK6pFH;-$9*euWMcQ3s1+-%B&j%oeSGbAk$
zSwr)G|NnMmE^{eYD<;-l$nW@|Sf!&iHto7bz3}%a!+3Rszy2)RbU_1%Og`}c3>vV#
zD4u?pz~_8)|GlOck5VV}1Ctdb?=Xz#J7JQP?GA2s?E}|purNaq&glT$nm!X{p)BKs
zkmLZHkei!pm`9{-u1aP6rmu}n3o4#_dlgUtuy{fHZt>|VYe&Z%pB3|CpZ!qSDQdmu
zh2Ihv0AyqQs%9-MS?p1X#ljsf<n(Ye5Bd{7?_ZzMdC}vZ3_4P6V1tUcK0n~`YD<Db
z)&$Vpz=e`LYBj{)F#2y%2M}d9kT0c24o||+H_ZNo6?%O<Ypr$|RUON-(;erv`-W*i
zef=A`DJ#lH#@2D}H565?IfOV#%tLbAONMWiUkMq@G7A9X$<zafE8eohp`mBzV-o%^
zGr7J?v<m@4%_b$f4;ebqJWZz})%i7{^3wMxn_je{z)e<~(iO^@&pm89Ir*GLD*ps3
zSh!s^P$Ar(SWtTQ6x9_@Pua#B9v*(Imkr1_(q`I)Fj`0oqpVcmK-W>gAjf+Do)5`K
zTs+c@@+E!X!MGKg<+T;`i?XJTX^%Cp(O~6}(C6U^WhWWiUKd9VXmPgGuwVis;&QsE
zHYLNazgqOSowISgl9%SQ&dUQtIwEdFP9fQ2b69elz;8h>8fcVVhL0vz2}S)>^2A8Y
zie#=7U?k>T-)RU1D7sM`VUF>upDh!O_7QuphdP;iO;W7f?cDzRiGR!=zr6!dkqt&I
z9&Su2KsgWM^X$d*4<q@gdbBvG`RhaEgHglE>Bl7PVLv|ja>!cOoSr(e^ji#KvQmP3
z+L<s|F@=!e`4Kg-&zLoZ^H+0^OQ1$e%#1bF77a|Q!Q&N=w;I3#-CUnn*7tYMc-BkF
zN_wUSS%4*5=v<ukQjbO7g8U50k8En@i2jhg5~jq5ej_W7dN1kMQ+>i|tsRGjt^tFp
zc%l?y<X}L5be7}WHsi%TSu7udK}THa!~uys4PIyNJMFu7?JDvjS-l6MKm@V$K;*@X
z3ROT}bRr(BnTF*fDFlfj+N<t!(kaWGML1l%y4H*tv?*J}d01Psyj0c4m)@B^W5)bG
zHY~^#+Y^|AWSwC<cnqE;@t?9j7<_QA@5@yw-m6jPN`EOt&=KE8Q<@El5W+I2k8nza
zO^kbAuXouILIO2LvJ2=v>9Dh9ng=D6<@GXa0j^L9XZJ{-D9e6mlVn2&Ev>Blu3Me2
zxBZu24kMU=`x1A7bjbS9NfM$eiY6q#b1SGYzetZG#LdW&&&P+|pD#%3)APrddmri5
zoUpjW=U;*a532KV3`qQi;jP8Q+sS4T>Yw{}CwTs}70AhDduIZhwVd!7Ns<WC5!TYK
zEDE&wG!K62afCzT1#2mn|A-jyjM6mZNd%oNJX%Fw!T@H0g)1dasu_f!AV*e+A4)Ju
zg}JqLzDJ&<ptV2@Uvo^xC!lK=p@R_N3m1kVMqB7v!NgKW3pJRv1+-`>m1%*p@=aC$
zMRLu160oXDPj=`sK`fJdeV$k;j<w5P>hlV?6<a!Dal<391Bh!T3Uo?iT-Nb#{aF6A
zVWc-Zd8q%K%ownxqI0^8+#RuKfD2J@>+2GhM<?Iu94$S7!ZIs3>N5;DGIYvBpE+~m
znK>ruxAPr4{sRC}<Rj_@h?Iq>x$7dUOT=A*^n=dB_|ZO^_2#9jD|{ph6j{X)?=Ak~
z!MC$1TH4pIp)9>u-JYv&o-52xR@eU*`chcX!eu5Ke*E##sdoV3hG7)o!83aEyMu#I
z9@6e(wCwKqJr}1w&^@1g?MeP;vKtp2xjdvI6C-*AaN%(M=Zic>PAl)QPX~E*Y;k%-
z6({V63RT+}yNIE-qq(>2P<%i2sJhy7(>TZ2hAaA-3+d!h_eNyh(xqWdoifWChB~Cb
z2+p!_$Yh4a&;;!kpTG?;i0+k2ARB@uBmu!7qwc`>-#;<?<3*qD(vBZh7+G+u3_gh2
z%y5L^h!NMLqYSppo0aHR@ap6Z3M+^S3<wyo@yzXy=K@;#-yIR@nwcb-I~1&I(E_pn
zucM|umH;ki#_tm}s_*LR&x8K(_`EV6<<^G^;eXRyMxbp&Mz_E{x-$On={2KCi_Iw1
z)Q~&vvoy-dAP#hi5uR>k6^6uPWbbL7{VQ1DvnD4=Tpt_qLJl1=RW`eIE=1>o(4j3k
zJ+yFVbnkAre*A|~ynL9RUaln0Gr<_8zI2v6NpNdGxVuN27@TJiRRP@XS(Mb%yhmS4
z^-ymPIp?;qbo@mP_qpGUeN_0q!1RJnT=`d|UB0ahNE}#Z1t11Hkb3dbHvJOjfBddQ
zV_VFaA(7WH@A;Byj*4`ml2)doGKNh>ifqL0V&io21_}m6)KSPnrk=LJ+9EGgO2`^~
zbWl2Ts94)bOgwYHqCs}kHBaiFJyTEWnK9j*UagyTnuZbK3ervwa=OIH17a!TypSf6
zKCQ_`_xo_k(xuy=N-67)VSi-=%iL?4TyyjoN!{WsB}!C=TbWYY_FXxa+f7n<G1Yjy
z08kvSfN(jC+){-860{*pFo3DUK~NvaQWz94)9vjSr7e>A87P3$jEwwXhB9Z6E72sW
z-IC-&w8<N&pa~KoJ#B-;GNC^ZaF$iWQ<0jIONcs2XLzFjo+~}cf#4t&WSoB@sfFt7
z4N0^n-_P~x3nwE@4;Gh%;71sJ>Jx6AA34WE$To69#RgxX1jjBi5ZNoef-2gMS`pbj
zQ#1||ufw(ret|^afVtP<+A70JXL!GAVm3=&1V*PP5QP=VRHC_(@DnUAA`}at(_q1l
zTkJH;W<edex-jw=&Y!;uSt|K5k{KX*VZGwVDI{lu5mRm*5^$VgG=xe=dmt;wafqk<
zwC~$AQ5DekP@cDQ*dmY?OcXuEKsH`N64QjgK5|vT+nv-q?$&Nb2t53GKIxluuwKve
zqHGR<-WolU9cImBt2d=G9PPjvci9-MYBHgDEVoL@dxvq6vN`h6MxygL{(GfLs9Ijy
z_%ErjycsdcsA9kG_37%T)osQE0U&K?1an0v0Xn68)Vy%kFU9G>$hW(7vsheGSGL~q
zU$CK?&i!9fo{QXMzoJvGfJIfg3{vu+^^gC1Mb8c!8+9IU^)$oTr@l)V+DfbJj2a(i
z(U!U9+(lb6LY>yTzZD>s%xLm`{690J31jR0vL+zocPD*k+Q#88iubLq`3VjUBGl`2
zRmc0|H4>BmgtN6^T{B(mg*h>W)!xV=<^N{0LmVixWHt^@?K`4}=D`~w%DiU$LQXgd
zOinCjAxuTVA%uPL9=nF?P($&gZtgcGJ<9at<le8C*ASeTsynS^_m1I>y7o8QYhF_S
z2FeF*1$qloC%@=r`TC31ua4^T(Z%uJ6*ppJL$~R$v=B5(AlA^j4m=!a3L-tz<=JA?
z5j8o!PmERZoWZ=<jm<?QUU5w**?qG9!m#G10#3<tGdc5s*d;#|!A$hY`0m}iV>Gkk
zC=Ec`BQ2r$iDA9#>}${TaBxBqn_Q7lAl7=E&?amzl31|y^~#X@d%eHm=?c*Hd3G;=
zi3tEd=Dlk>>P?k$0{k;2Rjle6C5u*DL02!-6Ad<oI+3dYfIJ{{G_iR2e;KH}@`?+4
zQfcZij_A_7kyQkvaTMX%l&Z>_&^4nopM6s@P;;X@9s3Kn8d*`|+mt4dCK<2NSP-}j
ze(6;nl_(l6Bu53AwB-L*cvZR1DX+~%kFvg9JPkIU`4tqLsIjTE1OZ?{%Q^eXzwco`
z1wLW)^GBa;a2IVDnX{<6r@|9RFebZUO`cqLWV6+bRjZPU3?ylq8YrNk#NzX|_$g7&
zizEwhZTQn-7GaCDQ>hIwC9=>!6YmwVCBq(Gc)j4C%vdAb%8a41MUW4YJ(b8w6@^87
zfh?ynMTIY6hp4WERzle0jJ}pJeF_w7<?%ajZRm90Wl=|w{!^gr*F2lY3GtA)0Frs5
z1uCOxBtETOnC00pP9riL8;lWMHlg(#xgt_~Ayx`_M%7E`1)&h+g{gjZG6bP<QEAm}
z8{ao(*O5PIQF_?NPyu4j1!j?0G!KXvk~^-ykzgjdt1=;3Knp*{t+<eZiHb87GM<}#
zC##P>pStKfjTXSu?DNmpi9(Po#bsp0xPzpya-oq2g)(i@tK{2G`wvF6XKh?IGba;q
z(VRX`1&$l-8g;PB^1qB(nbr&T00-df(`BO>8N{mLC98vFT*i%e4GdPcT)C7$I7o@y
zW8$qE+SN_bQGg|BEFgtgr2I9YXD+jyBuBgatyp>aP*};sUazEvZ+TP|+aX;k6%iv6
zWaLk_L<Cx+;>a|he^FwW?tde0JSi>IcW~jO&tjiJWgt{=LQhhLNF=_9bT(O&a%}ET
zd>^=4v*sT;O*|oC0RGwnI5P&vJ`tW3SzS!upO(b`H~PGS7rWN{NfI-ho}QnMaaCa+
z?6>gnWYdx(bf@JI12gERcsTMNd8ppw?(nel$kPm1t^XXb%%K1?N9l!eC%omw5pwT#
zP1WI88Xvi%7@0HQyPElIbzsAu^sAcX%{RAdbU1{#;bhn$Y2+BFtiC*DjNMEbf|kde
zv&GN{1quSj-I;zR48nrDBj<k0<rmG&lxcqsK??f<jKKW!`xi#tK`;hV9AwB*P9%f}
zih25`6(bfG4J}~Br%QoQx7?LQOwoppdfhkH&D_}F=kK)1Kixq+enpKL?zzW<QP?eS
zhUeT3!ResNrpLydPSP>>?cBM7?@~O7Evu-^-QDsvXGUdKw)g?p?u+AjHGcD}UNsxW
zy!c!qEU`*4>*ccdP1;cqDLGt!Sd_i<jyMps&)<5>`eDiJ0&cIs#J?$k?3J?PH>CIz
ze92p&HzLPK{w4Db9azNh)N*{5+7HReFsy)$0_!ABV2hBHvazshSN*3`63#kpu16SC
zSbE<-a$54THts$>%Um*hmNoQ@2-`P#TKYA&b($t8>MWjGR8Sn`!YsyNl@<|-7x;pX
zLR1iI@H?0`)MuemnM|N*+{smBWc0E-9q&7|%E|M7d|9t+SK>W<;0F`_yK!&-`E}27
zt?&GQrJZ?PjceQYSH^8t=FG{QAsRFYTPX9aB#{&n8fY@dHWVpiD$202Q7V;`29%;f
zrlgT-g;dl|lIs1QVekFk&vW0;JFGwUXRFp)*L7a!c?`eb?>LUS+gl!NhHtNY@Hkuk
zbg*-n#=l-q04bPY8g!mrMIuYIm?8Y;zCKQRJVVhY6uxT_Tred+i)+VLQ)%()+g7Ib
z>3deq`Pf^7>@Fn@x8QjJeIcsLeW5VCGlJVKxI`iBn>VJV*%1o_HI?E3h_z?Y%Ub2!
zKxsgTYEGj3(ux`tsa8(czZ90&-+)Gm#w8Tdnm^qzx;rx03z$znZUs7r090565Q(pS
zAeO?YJ-E3sPsh9F!tf5sw_hMBtinIOx=E@dnz#zNPSw2*6J8qI(tuWs_EobYOBONq
zPXTABpkzFSd{%$SmX~F(LTBI|x4ibZE2>+2MH*$&Zf&ZjQP!b9{8=;G=%?!PKSPtH
zQ_uwqiLU_69;T*-y^DmVAQm5}FNojJZB%kLj)B6UW-dEQ*In2$fVG!Cy^ZXp@2DQP
z+G4!?M^6%>_!Mn&D#3O@d{()deEStJHs~{yxIU(COZToBd+jN532AkRqa&KEiPld-
z48#&#tA0ln4R=(9+U8~Q=6@!ac~MTwZrdeEi@BGKr_VOmci*7fPFRJ?e^`ab!(b`J
zup0gOxGy}i0FN7k9R@H0(vXd;==4P7k61;)8<8PV&T<*<OY|Z7uG5)I&-U!>`fX3v
zfB-su{&0m;FWv&PjOPs#ZJ}toN0CH&K$z$cC6vcP+5_1tCLd8=BGN!Ym_YOlFOA32
zY*IWM%5ILK-xftAlnA|jVYKnNAge%3t@5ICq<Xt1(D2z)2-5%{gclWkwl_o+brp&t
zmM<Os!WD9VY=2a4^^^a?zpB;N-996|7806sS^o6b8kJ(LZRh`rB%gnK(=FyP-_Swi
z+~E46@CUUzb(R-M#7oY;GO_v}1g;ysKaor?U=|-*fWi1F(F_uAEMgmcJehvz%6l!m
zeWQ5%j3ax@beF2L4K-&)ekceq`0BM(*<lInZ=Ly{^4F9vG_?|w;zb=_lw$~S=>P;0
zqTvXt!gCbnjKam!<87%RC_diFhNy*E6RpH>0;;uj>)eEy*fe0>n$+T_KIMNNx<C0v
zW3DAxQP9YX)&nI!hHN-(Hwz9i)TZd+)P1Y}CFSc?c3Gu+TLGjASo7V(9??|%iBi`K
z!ym|ulS36P!vCNdqJ=ZWdD)Fogm0Yfu(h>;4PQs?t3K@Jwh#dOp5y7S+r_xy+^Xkm
zA8Nw3zV{t9Y}g;U(IwB*m#XTCF6*M~ZC{3JQKiS}YTxL2KmFXt5ab%^<cT8^WPkQG
z%)jvSw~|=>pO09w^XSij>|gY`KaKUj%}9R+%>RAW1t0uR{!h@=|B3(mX;#qr{)EBS
zOxzyV5VpD1p%>~OLdi>G+X?voOQ!2v&D&(1;^k-fGbfdBnOgK$oZ#%0u)C*IOsm#+
z{J(eN%Z~cbl;g*-*6%(qr$R>Jcl_Tdj}Nbr>^$4A-tN0dTcw5gqQ=*MyN){#0r5ij
z?F*!cMI;R0ySP{xue(ZtM7;u<RlIfB=Ajdh{^Z$WrviGBp#aWYqBEs8y$Z#s6ft=e
z{Uwys?dy-My5M<H2cvg@M0~9HbNT-#N${?-s(&So`iR8@8L9tdcwp4sfF*wIe(nRm
zq!w+v_tS^$dP4pAt6TE<#jZbhz^pA*0mdNHPwmNGMJ6G7&XG1E-00G`Zy*-fsaWgZ
z3S?ix*pqS-<z~EI_BS<VoC4JJv}p%zKaj6sNWO%`1s+3wndQGN*HK3%^dIT8QTl?0
zA#4NU7nRQ<DTG3~?$q$=)zUl6of8Q!S*mEM#AQ8;Ecl9Y<FOc{fj3Xe$uaKVzrRo`
z+9Au>SKwWVV*l?IqON3sK6$F>AQpz8sv3&tnP?lec=7MRUt*viG^^-7#(*Cy^$gJq
zU5E{6fPtj@IcbcjtQCbbKxvu`r%m~CyX86Ko=O>4ob|GwL&J!^(^xPuQ3|jxUFe#+
zEq6ePH%ZK9?v;9+@r`(gAw>Qp--~fHMBi03b{>jy28K|r;Qu{^i`@4@wYvU4rf^x;
zE14fKm#-Bij`UnXzzY)lt0*lbM;7hpyg#MX_<SnZB&dIj&gq-&j7~P5l!@M>y=DrV
z4$M1HUv%Q5I_Gm5;-m@RKDtl|jv&IqipgDRVuYcXmaevPCE41KQVckmD?lNh0nLKx
zGKd8BD5FXI3DmBFey0RDQy1|SX6+A%h<CJVW%$?G3&SJlJ~-6#65)HKC2W$4<qIVW
zs)pemSWgOB47M{<Qs5Y(p8?z8Mx+`wTcTgWC2|tOzzEurlk$sp_B6G)_q7iVZl(*x
zCCVZ2TSf=wys72t?}_3`k@H5}T9#R0NJu0)zR`xC9ep$UEZRXKB@+E^PzWJtGAMS0
z{83r^L~%U#i(?B9Q?Ig3aNj4+n_u%U7lr7;i_|SAp!B_O(2s;|)(+weMWOyxE9FPL
z4I|+hZyNO5T`+f{+a;2|_Bue3veqBWmEf{PY|Z_tS@tHD7bP)@nR_BVp9QU&hMIPY
zh!2E3Er9WY9*S_M_v8bndon~q00gY8n3Bx(vloLRwg|3kQ+U~F(M5>lQ6yQa*~b-e
z>7tc`AO=w_EiI}SD!f+q1F!r@#KdF~RzeiVdI~~=FmOC{vKp6KA|EjRcmFGA$8Ohd
z7Bz9KnS7fKFYgQ*^5<_kZ6uHF)GHb~Dc0}6)Mgt%ISB_k%bV6Eu=(4ou9}GM6>2i)
z6h*I2iq6muU7)%2IS!m*E$3*Rz5Bl=ZbHBhBdM;HKc2z6Bv+Sc3M$enG3f_A4$ZtR
zg*4)y1e`tL8w{%$GHo2I>2$Mh>dxEem3md7>NR{4lcu)eU0p#zL8(qpeOTep2zN0C
ze6E-a)zdP_;Mz3nWzM~(E|_5P&7slvJ~sOa-|p<W0t~W!)AZlIz3#B%F}bAB2yw;Y
zgQ33{GKYy;#A`C=?TT*g1A@>GZsS?<thp;uw`Hz^qA2b21Q|Hu^gyU1)MUy5P6(Zu
z2p8jI@Bh6GAQJmXo%87i<#$NMuGeN%sh+^G0Lh4&a+DXB{rm(W*S>GzAk|WK;CG9^
zb8@dqO(bjDlY5~H`~x{C1o#t{tEwSO3?05&-B2*6`=pNQ*&{kniECc1G~mXQ`9rPq
z#?DGox_;wY@<pYS*$)GI_Wii(^xAc01J|wn)M|d5ai@zj+cmeIlh`i4m9)C$gh7{j
zbnZCmW7VyIfHt|sQE3rr&T<dD$5t~qttk!OR3#jdK49%l>ocb;rQa1a-li;T`bCEH
zzIQ!xfsD77VZ(^4I^mJ(A*FP@NZqnt<9Jq$?R@#yq)#+z=-92>G}h?ti|aixwsZ-9
zT~@YXY|`1I8HQye20E~5Xo2+m{HkST+4{P=if!QD2ag_o1e-Q<P$%d&%HsEOT#j+5
zDj<XwTr*Inw~qd-Sp!A$viG6zCePl~k|F*QyKF=330OL#ttMQjgohS<Z6K5GS{5qR
z_!!q{?7-(9aCKb;TXEAW>%}>y^OlBHn-_0B;arj$6WtNmq4AaU8_KxH=ccqla${+o
zlbU*%Y64~Ww%9Q;qHPg`!d*CFtHTlK*~mvTBkcp@9gbL!dl@ru&>$N;i>oE=gpQ3c
zI)TR5mJ-&h&a|+K!w4t^WM*a2bbA_ilR(a~9((XYNy6u^4yoHvnDK4U(F1_QXqm5h
zMjrWe`qk|dPg$qlrTM4?czNHge|bExa<irBw)v6!_jevi!vn>UBM)OCl#v~{B3Ctu
zNkus-gFM2k<||a?sj(LAS@`ux_Rnouzi?q^2(*Q?CI_bT4zze-@2j#_4!Htt08SOt
zlA~J-mpIenE55$HdFHFAt-NWOahTo$y9lsldL-olMZTKZ)xD?sVfZ)%63GM&4$wxs
z+%CLDHuLi4+D??EcSpr(U-w~sl6y_*hYy#R-_fLCuh)d_K1gRzXR&~c<QX6V5ZgLl
zJuoC&>s8!vB;<`R$Ga>zw<XChF5Y+YWH$upO{nklr#GSQ)TwfmntB|uv$b7?qXFD&
z!P%fU({vu~r;3WZX=xpR{c%-87k@ar?I2A&E`TPo!}6ul1l+zO5b+g*AO%wEwmsF7
zhlLT;r6Y&rl+Hk5wD$QQU62Ocx_vtsI9D{X+&qVFN~csUyP`Cr<KrjzwkP1OOdYrj
z_w3^B-4j@P+nM7hhFq?h{Ohk0n)QwZs96Q3tPOy3Y?MhK@*Fxs#`44&rK-52u4+lZ
za*GK2V?Q3LtgMIjQ|d>y^9x)z!p@akNiR*i{p@Vms?{ThZ@S7NdmcG5aqirX{QjCj
zP!pH9SOBtLLEYm6mP=Z;Y{_mHAK$`d7m|_?pa{{vzw4;wGgqwW$97gJ|FGAsuo426
zy2b*kxD+%sPa}l1DM)CCTJaw8d#ixKLx$`J!BZyT-QG}VPtG<O(FxrFq{IqR-`MKu
z+ml&nXVkjXKdMsevKnlG)v}@QESCNu1B)*&ydp<Wu4Q+ex+Lb5C{_P9U1k5z*Ivv)
z<n0)eZzY2p{iapZAo38sI#+8c8w8u_51Ml(nSO`YUvZ?Y%KC(cgq-JM?A|_>Nxb|T
z_2?!Wn<itX_kH=Mv@`|fcSx6$?(TC=6x^hv<D)QI$bLPO)}f=k{B#h@lv*a3ID1#q
zSNk;Ri=lB^N{N!=oL0+O>?WO;2RcTlW{fbgfK-%~jfkOTYiGM(6zpex|NXPbW$$*k
zpCs&(u5N4P=YKsXS=pf%{(frWUklHD|Fc4w;srsIC++LU4vx6cEU@v06Pu2_tE^^H
zTrxFhtM43L-3vn&9B3iFvGL{c>vZvTKVJXkU#i~?qi*4@X9$xTFQ+hI=+9Hq;lexh
z-W8jpA8+-YrK{Vg^Yo7tBJoOWdc{8CN%P*j`CsD&yTgCEoS${uZ}JbzHZ)k?Ax-IR
zM_pZJ!@l@_e~tGz^oJR4{88(F`Pbj~A92XGIC}Woj=|S$ii+E4Hj_=V_{X_bn6LNO
z^|1`|^{(i;==HZLSn2-PV(RLa4fs#*_?P#$2<*I+Ejm(MaqH05(bvUyHohWeO#e9D
zfAav<aeXbkz1(dNG~PY`F>{xTQcy6ByJWiW?{5v(|GAp~-Q6Bv!}7Gb_&!4H0R@E$
zNyMic)5N1xb8vE!>)Ur@NJ?g=7dX)#@Eb2GU{;OpUZHXOLso?CVC;(-b36XT;Ksgr
zn-?>I63D?C8a=votMi^waiZ~B`hRdYuF*6JAN0Ipv(ERc51|;vR7Kv}*7hC|<GE7G
z(zr4%BMu9?|KP!M4xipM@BFTmUYx5`<n?FF82)&i+tdM5&j(Fhl{kIbvR({Y)rJsi
z-uqyo=ReNZEa%T3L?`>M4aRqzjsuv|r?d;E?UBIJAFrG~k`JU);Ry6F^B?;s8qeLm
zr16pSADatU!aaNU%8W+yN{OS)G{uVi`TzL+H3$j`eOEVO5t<yHS{QdxK1#uJgiaT&
z&M&hy9VVBBYHkjfR=VX>IO@nIwfS1Bhg*MgJ#x#_r%y+`q8AP4atI=>*Fe`;#WN@l
zM0X*uu;e;a@rMEfM*-$ap&Y_a4^UQ?MKK1sX~5K!ny=6br_`+Ev9#mWH|AcTJ5m$~
zsx}0sVS;nr*UiH8P{=sWR@#<avXBN5{-h``d2F>*2j1XrMN$Ayb`O7OaBEY4$eN2}
z7Sio>u6~Wmy8xzVfQE(_KCzNf#>z;#=(Qi>Hek@8-)Opb>bpy3BF+lzx`V^Z?SVm%
zu)vf1-P{y;A01)Rv5U;gnH{<=EKMbep0i}7UdpRlNC?^Prl)&=riH*G*`7tbYM$zo
zge5-U?#?(LSNgqvgb$s6PPdh-Mo4Y8V!1d6nR2hYd*kWPZ-`DE+##&t<sIL^%DvK(
z^^H!lIsuO$*<@#Tf$&IL!gSQZ=qQR%-q2uYFW=RvTek-oNkeXy?EYRxMz@_f!j+@D
z|AI9qFALVY*+j`Cy4U&iMA3jaT8+X`iWYB#hhKs<>O5+>4=Z^CmPE*$P^qlyQo>W{
z&g4rl6ua_VUy=vDW*V&CxUFAC1EH!^A0oCwPx7UvMwuszli20vQanZwi*5U8cb5SF
zfu>Kivr{2UmB+2VfrueUdNJ&<7&Jfy-~8yUj&$^xG;?NK@=^;RRCZ*#ze|x+=;Y+&
z<<~8@Br3tdb$~WEBX$;U5+#sf_R*Z;P8~Z=#B++2PB<ZU)9?IBYAtrx*|UAwIrs0~
zbHPyn0)HNjijqh+77eepy?t7`?NR5+jrKUnzROmlZjGiu!BeoUy3?jTrEE+>&&kyu
zr+TZd>&&P>GxU;wmSm)D-A0C?T~m0pt-Q*9lv?d1E^!m_xPTYEs2El-Tsy|=l+HFy
zabWqYiBq2_<74?K48f!bo1&d=V`M3e9>_3tyyENPVhh<yQtB_%JBSGs8ZTI8s{OpQ
z9mg;MU6wJbU8|e5XfZ#?bjtSHQg!y->J9FP4y`6mTlmMq17IKgsJYn((=%gMx2CgR
z|9<^;^7)42BatVb1bN-fZd+ktu`kKYlmwdTb{?b?bSCM+prK}FA3C;g*X|;?2RB^+
zUDC&b)6OGxAu9^z7*$sJXrG0NeN$5M2!Sfos#Uww=)utkY?zDR=jz&q&Sw`fCDIHt
zgv`9PDa{ywmP6{Y3v-8hw#p^30<xIn;4qG5ywCPWnDsGnfV)B6pts8p9Y3jz>$>wC
z7p!GZyJT5eM@i-|nH4JXDJJ3TGd+=68{C{Wb!wBxh}c|9g#886xp48~$}wKaH6Gkt
zM>)B>oUIxv?vm?7utmY?Bcw&EXUj#`ZArLPj@V$u?Bz|44Vkx_MqF~lk3>o9kLOc+
z^!pT>ZcJD7MUr;|AR6**$b#S9;>W}MV+mR?htRczR%p%h0l!oS+sk|?E$uPhNrf3l
zSkO)6c&;{`C74}v`oKxc34CEHo1WeFxd#&Ck;EaY?mK8?LXg)()9y0g<;Tw?J{0b~
zaoTV;Ed~X+H=aSRPoE95%ght9h)Zd#lSb^78Ra;ZxHQvggH3toh-T#=1;4azt$^iW
z4n=-ZQ9Kk>0x5}zUpJzt3+r?(qWADotG+vak8_v&-aGBfusjo3`L1x+-9SQlh9?>>
zqg(Mko&cgY!oP0Z&4^dqwr(A_<wJMTGZsuov<sDn0er=+JB<05hlv8D?*ZPRVG#^o
zBN4R-{m%ckfwY2DC)~?{C|3<K@OJ~{61xoQBiBn!*B_sk_~*mnfd&JI;tOg(4u{S=
z*9^0IE2UOgOkG{!)fczQg?4{?x9Gx8Ex=Rhcr=6gXa-gz?vgg!$k8`s^1~vNqe*5X
z-K{N43nCqL8q^T_K^jFIONgQC+);ScU>c2FZ5DXFcu@6Xj~O3lpBlXjDqU2gEe%io
z@}-!{R<F@3oA9OD2ssSv)4Z0fDq7vP(Xe(nw_7`Z(S;#T2cJ~Dgn^mNSKHd5@~`ZR
z-gIpMJ8r$9s4WF@{Zw78ezIR8McduA9Y$l33h})pxE?1?OvTv=fk4Fid!PX_!85Nl
zIW}z3LunOEcj20LJ5yZZXm|drlanT9Hrb|fIn0dAm%8e_>#u?}JCCppw=#csa3<F0
zLw3YbNR1oF>X-#};LxE^!;H@x&ZNZ@&(zbC+?ZZw@o=Wn*>To)|CrVCkD?fb<N1f&
z-HY+Dj5^QuG1%*7nnzNuBa$f6-U2{FG$OCE4|C8ux6qumRJX$Oeel8kdcU`<+PUOT
zOjnO{es~bJ^AXYw4k`Q597C~{hKiCXs;-$tm-6Ne;GQ!2XBwErBaUT*wxmXk1TEfY
z0?|s-F*ZsX$Gv+cm#%wI@0<~QJuIf7UMtHQhrDv*M)$ZCqfNAS)N51edOtpX-lf^S
zVR3$oiJ3r?rdyGbMgw!pr8U_!z?JC{Sis=f4OtoNoA;;Jl}W|@<q&u6j7Y4$KK+D}
zJ_Ed7hHvZ(hjins)x1Xgkytk)=8)592el_9N{x@y)961y3QIqz0AjYfIQ_T+HW(DD
zlbl==vtvyXMca7^y!w$MALiIUK%{VP!}{e9vt@dA>a=R!v@eT+CU%w<GBaF3?EKsH
zh(A2eoSE&~bwqLoz`+c{G#2l7tZ&!4u}34qA6_l~->{3zdh6#Zy+62cw>=!(#DoqV
zjmTpkf(eYXesjO~(W{LI_(<)NEAyvSScko;8FzC3{=#Et&%SpR(XQETo1)eb^h#(m
z%rSw|651;=$;*|nCLJP^xgP60@<YSJEpQDpnRmv<y+OH&Xi(v5EJi94G-<BSi}s}p
zknp>%5wxhD%zCcLiyjaCcWZD-2*Cj64j&vuKG_Fx{8b(Ppot#=*kUUP;vFfa;SaYi
z_EJ4KYfvYw;4LWof%g3iG#ZSS>zqY(>Or@TO&{|ovFpvAM}w9?r6&-iSm`uPGMO?g
z&fC#u1@W_0m__~ETbo|r9p7{KDAQlsv>Ds&Q8dLma`c{jO>7-inW`ntllP5TX>4D#
z{Exyv(-t@EIQzmO5dQGhx2iLbGAgWcr59@rw(clMc(>I%E9u?v%8ceUQGJg^w6RHf
z{>KQV8xb3g?YGx+(GiuE#hKS9JZfE0d|8^B=O}~cGpn4nc?;o?4?&dSvPTWYU()j0
zTypP~8t!;<eV)aqI+u!&nircV4B62w_qG&?kTff5Q$tjBV)=xO=2d-`q}9gNgg>hb
zHc!-ZXxB%3oXM&q-xiV5qC<WeYT1o6HgufJTZ(d$&y4SxIL@Fn4-h&HK7=RNhJebA
zV!}Vt2>u-imfk(P(g|pNOtHy^+$ep53CYH&PYHJnPG^z{vN-4Q8X&+faI)PK^&Y04
zQg%W0{t52kb!n;9=5iM7Q)MOLPamlz52ac2=C2D2?;}#&>9)rKf3B~mmk)#b?3&@z
z><L4M9w3rzf`c8AkXBk5!GOB>YNpn%0B!wJTN^?-u&wdRozySkwyrQ4Lg)3?vqR=L
zwj*z9<|yw6L_H<OS$XmD`<oygME6#2E3a0$<ZZcwoRTabwAPf!$|j8kaEX}xG2#Aw
zHerSK;mb!}R?nC;t)GJ9%yR3C{n$7|;*9Q7j2%Q{dSO~CEAKshdK9AC7N%=^eT@bA
z4`hMb(X$lf`4RNeL2T8=jdO^uuYDCr3rn~g6~MfJWg2H3r9Gw}8xqH6)6>`Yr^gC+
zlJ``Vjvae|8<e3nkp8b>loT6z;FvL<1e*omL!t_7(W}^`qyyk_=@esU=j1pta^NNL
zCS5!ptQg&~M~}5ZD@Q9T?tvSmCFl%ht&znQkQuJTJBe!`N%0kE91-x;!@bt(B##`N
z_8>R+45^rGr%v}Mrd#vs=1_u!gp`y+Y`1h`OCmhzgjU!Q3&`1C<DHI)zwXh)YjypM
ziX|_)_`CWgE2KBqc#)GZ7lBaZZId5qTj6qsX!MS$|2iB{kL2tI+EIcRUFLa^Mn0@K
zS^D|&=e9k@G)sJ$Qi2Z*$%9}!aQN`=>_|eqN_;B>z?>U2tOk;|OM9{TD&ki@L2MV{
zj##dC!g^vGu>@jquR8nq`l7Ru5c0)y3q&3#&>~b)nPh`7-1Tdu1SRQU&?S-a5Jge4
z?t1m=a9U@q08j(F+Rd<DSz`H5SG;}br5If+@NJ&-G!!2-PN3>+(0)Y8BuWg!+!ePX
zpHbueQEQkk3GA$-wP3mV_%>qNA>ngtMD9V3rg?1PQ^1&kQ5CJ>JZ=qMRWzlo2--9>
zGR=+X67@XH5pSQQGi};~Wqog3gU}N7o_rvSCDk6sta^8*UryTRv8p(A+mgrq&5Vj3
zrjY{J7opfOLvr`wRj4bKVmJNn$qwy&@R`ng=>ASUd&2OwCaVlu*TC34nac)2ZyN2_
zWm=3(k+tRvBPM7=N=fA{g&*<^udlD&Xr6p(7aE)PWpYWo6c(N9U%4;_?IAM{gMuXF
zmIY_Nhrr+Q4DBIQRHS0z9njYVg^!v)bhwA}A9{6blddhQAX8o%ifB<pJB;7)y`B^F
zkC+d1>OBhJUqN=?W(4*{YlIEy*}S)>p8xy{a#8dG70D2w*#RV~-htjT2MN6J$<wE;
z+P1x^8UK+;hJE<#T$t5{2M<fPH9AuM)8QNtAE1ahcl!?<@aG`@_{{c(P4%Wve{)DZ
zH=or3=$jk5ZV{S_R0`%le)_bMa*M@DqPbj?(e&Nm>^Lu3m6P)GgKZkp`&%m5Oc-v{
zYWYss9}!tFMsg-TfT(~pMHUp*`@r*Sa#0aIN;Z@?B6om!wXR?+yv_43EX^b89Qa#y
z+QEhG{sFJGJRq=`pRc{g{?GsAn`Aboi(N<FI^Tg3-d#Kmfej-WpJrn-2Aui=9#>%G
z&|O&Dd5>3i9da~Dm;6wMhZviY5x#RkV<O{m?$_@L=5Y!$b-sO^jC9BDK{Pa&1%EtF
zORIRYS#{r|Nqg>WOK5qCv5vnYacVP8fqlfo9QGY*w7G~FSy~tn9<ENHArk0s)WP){
zw`nIcNWD9G;hKI&V$}Um!tS<)jYi@^VtRYA5J^E4T`UdX&@Pr@&{*t$5Yc~THAO)F
z;JA{lR7AV=8Lyf5Cdy<@%kKbY_V7krQSUHg*N&4Bv|t7i^EQHKmYItp5ibqfLFBc|
zpnZKlY@ItqfRpoYwX0E$_M4m-_V{a=z#$7mT9%kBVf8!*^B>8YtgzDFM<XS$G%fQ%
zTbAMV&m;_Ld@vRL<~8C?HrNPL$p_FhqdEc+V%cfqYzlwj2GfG`_y^scIyfMA9cmb?
z_+$Os_SNyBUPr0ii1BIET{efdzh^e)6KCWJxD3Z*7S`hJ(@PyhSKx>p99)r>;Wpy|
ziAZt;c|tt2TJ<xvB{gt7U-BC2TgH+pz+3o(i12(0%F0qHoP9{16;h=X?0Ne1L#kl_
zg<_vSKTZNwKvE1Naf>t2yL=8XKKKPq8GaKf+J_G_DNpMTc<|}dUW*Km#`9h1)`;Ed
z9fqkqB688h%U89~|JgoKSBLu|>=uLi!b_`3nJRZchcC|P-MhE4JUUXFJt8A#)QNZX
zG+o`f^pX-C-uUZP&&%@jV=>B8Ir+&MNUU*+2+-{EQS^->B`+X6P<yb0wyUrjCl<9j
zsC9hJIJ>tUDT{3@ru}FHxVSV0rxT^+YFVpHO+Qvv7LZV@Wy#<o%yRWd@US$_%eCX!
zJ^it>H~}C_lI~;8+wznw2u_5>q;8;>dp9{b)7fRwU*5RlZ;p~vh_IO8X0WbB@~Jq?
zL2$22(9_qJ-#JZ8T0zGvPamIT5(0tt_XEAaiJRi*NkvHJyK#lZi0+!lu2HEFBkp=>
zL4I+NHX%62-6o0fnY}^-6j=RcJCXlDZX7)y`XD|&{hHxeiX`3l?Q2QLxO|cpvX3B+
zGGX_@^fed6RV<5m^1hxk^aS#c{A@RWD}p_8)qIMqzCk}qUy537QbDitFB{KSVg04v
z!nAk}$AlvR&6PfA@K#y-D=yf9@iKJg>YtjL8s7UVQ!I&DD!FRm{GKCHJ<KVl1)ab=
z8zH$hY1wgYL=+{6r}Sv_WKMs+m6=FMvxfDXX2R3F9ZT#&U0J({2=*KwSV`OG&cqVG
z5$(ecmeZitlxSMwGtRn+ef5qlX=b4=40uVuWxa|ju>E9MIO_0Mt*pj)b=R~TLKqnz
z7q<&VfWiG}NEDJ5*YoEGKro3oK6?kiDv+s5`WiFvKf#v0uYqMqEM1Fv$WFWIBCund
z_^d2%#Mab-#bZ}?Qn(jUDxx8@h;;kdEjMs_M_@V1u~V#(mWVu=baeZPNR!giQb*b#
z$=o6-IiXBYmpCk<4GlaA+=-+@rk6GOvn(e|VUUqfVIR*<aofmoIyTZ+ekWj|d5?nY
z<YuuP#UN4~YAe(pEhD{D2E}ATIWx%x`qZ@){k6o8<Gnb^SOR(jB#FmoSM@WpZlnzK
zUi&waHHD||+!>GddkEus@-a9(tGy&{pRlb9!Z#S(#t_av@tf4NvVEozst)QXMAP?8
zwzeS}?tB**puJl$eW=16bI?0}3q|yjq7y!7TFBR(PtkXI7G<LwhFycAKKo@hU3Pyl
zQ}nYM=sT9$b~oV@1<#lMf|eB$u(FyzpU#2b4!s?NDc7brvTNwxMw<M>n#8)bOz>4u
zXqZoSfd>6kRaFsat^#2^VzDu_Mk7wfaGb|vvHo|Zbz~P9*<Ap%n;36G^WoW`tL%ef
zRQ>io4YM!y>o-8w+sJUB-`LeBmXCA(tTgq!Y|EDWvOedLxg41n=Feh;mpU`ZP$H5_
z>pJeUm_`C-xF-?tf)=JTaEnemkJ!lU>2OC2d=0P8<A3-M3bZQQXC&VaHXOPnpjUZR
zB@w2bAAl#-vtyZ(B_O%<a?&HWz7x)X?i%%evY7N6n&Q{>>y=2q)3V|Vk6f2)9WDm8
z*OWYa_DpM#L&MjSQT<l!qd?y#%=Fgwy4&&9u%CFj=-M(Z$i$f-&6pZfwv$wc_WUc@
z&C)U5#PTW3x4W$C!X*~!ZQ3-tg^SFRB^m4U3W|!R5P!=E+i#J+h5b(?^MD8a1ubw7
zWz%LL=>;Th%=&WS)O@7cXmv!1`h?Ysyw#sxui909tnF_*<nEHDNblG9)`Q=LKOaW`
z-mF<Oo^L#Hv_P&w5(K9SflCZ306TEDwATk4A8_IJOW*gS9@}@9o0+t<y>@MQvV3^M
z$Zj7g;p*78ulv!Xo$yIBeKorf5-~l&X-QG%vb_{Z<b<KkDxH`h@9X1JHZGS@Dow-z
zDc8BxkQN`mS1?#$6u!{*n6|sftbU0ls6BQ*V`ifvE8FqoX=UF&x}~BpDf9MVRn_BM
zbkoNW-mYQvfRczO0w5OBOGt1v+*HbQ%08+<A&|LBl3P93O49B;QbA!ikb-%pc%H|U
z4CSUxn?^8Wn*QSUW=kr>45GJ2q&WNf`g*CX^{lC-X@U80H4)+vG?|$xqJ<$aZOUQo
z<Mg&aQtgNI^cCxGl^c$L3RcFDIWRiYmfifuDPdYYBHk@omI^<vec?3&571|s(TrdQ
zu$;oI;}dm<rVfCT7fwtfkc6(KUi)0xfI}WBlcGEz^C(yQGm=9&&O6cyAcJhXg&<TU
zggoY936|Nnlu--HjFHQxowV8mc>0WpEv52Ub)sB@oAf83{7WGAAxiy&xW>v1V2h;1
zf~cuj!Nm6#s8>|*#&0M;z7G=w8Rtn%@DYjpW0FkD&sJ7;n$0;SYSTIq?wR#y>*4U|
zjIhnKbiW6TLAWuTRO?t^?7fkZTFCIt+0;1leMFBTByTvm$v{{Jq3gOx0Q2K=a&k<s
zK~*;GvAVVBbwOw|W!<`AxHv{jZTBaJq+Z4g<1Fd1x+7)%Isu>iGRzR4xd&<cN0ytu
z6Ny~+9%wmm!=__ABgs`$UO2FwuNg;nb#`{9R&zSK9)mkhswm?NQD5p#nevHFn>Pe7
zOy;(7^QJ8LHi^KKg>3~5lfv%{YeX9W2rSE7)$wZ{_q!_BLmbI4_!M_gDzPRMFKECj
z@s2?-a9FFc6e@BP@mQEMMYEU7CiS(~shiBcEi^gUZB!~I6ds~G9*r?H9-w#@nK8Nt
z<dVJB-H8a6hSE{zAlpJ2s2KbjK||OYdpkQb%B|@>CpH=3yfr_-)pEWEr)6x>g2zB<
zzYpuCj8(n*nZXl#NoAyKlG)cGuz%RkX<N5y)B5ekdyt4U8p+beI(!`rRXZV>BDWff
z=?c$m>BnbLuTnTf>~qJ_jDPjuSS=twf=c-f_2n9S8yizRCXRbIaB-(Tea;j)ca)8&
z{^?}dHIx<5J8~B0io8gh9Tu0H+lR4CKx_BG*};Vd3>}(eZlnH`7)oI3C<^@%{cwYf
z*-PcNG{9WhF9Onws(N7r0n!As*JP;b+?eg#<T`cYK+5Ij=JE!iZXvbE1X<<b-^<l0
z-WZ6ESMaL>ME0^c<In1H_!zKkD<D05&j#=0GjO!DI5DN2jDxdtL4N*DLP_v#Gk)-m
zR-)IIhCW<Z-I!a906mpxBUW6Tlw)*<UrN|KMKhKun0o)6a4A$@g?OdhzNy)>d+47$
zTBaQ_DQKg)XMLON<=cyA+jk@fC#=Q~3wzXgxOr{Ud<0Z9_!V5n_M|(c#sctUnWq8=
zcxi4PMrmC>bi2y>mq(b)vX9+iimg#KC(#s?BF2pZFC0NU0Eh!(p_GA<67uM~(90ky
zBG6Jbi2)>wP<+np7)zaJYGZTB0)&hPB5My5!XgM9ta#jDMfzlj?6LFMWs#u}h#1~?
z0DOROO6Kz~&-HJ@X<eBc<tRv?Z*4xK_{o0$HUDh2Kb?l#AuLX<Si0dwzbUR;eh55p
zS%w_CAtuq<2$HmEpw1)S%Q;zEMLKt$#K~~`?sxRNS6TOLKZBJXemiiD!^W0<09?4{
z`40S>5P!AmGziE~1YpcuQUL;Z0F?9P%MQ6xu&ekrD#1|IuF5bDWE<TNO@*=Bn115O
zkyyfe8dqJMV#ClRf0Q?#1Tb?MX-s07Li~P;l#u?c`7BL{yanvvhn&c6^wbiyv`fR*
zCyh<hzK7R)hs=x;qR6XNMGI$VaC?8J<E9u6W#Y6etPRJe2&IxxLkO;)H5I`zVPx63
zR7g^h&Vd%ea`wO^ogfNj9J(kAjE{?;bMp@^7?E!xql@AI_tH39hL}h+7g+97<<wxy
ztN?-SYCBa|S2CB`WWs8that<Ch$z1@6#1u66w#!RFm()}-TLfX9$1+mI&^`Rc+tix
zrK*Y%A-G~rPz|DKA+I7|h(%6!^j3ti$?WVD`3r0Qp|ntfA+rAZ=UM@k3WSE>6|>WZ
zu1%VNv-ikHg{k37R%=Wp3J^5#lq*f>f<-Hui4aCGEz%O)rWD6@%_#gZ&C5>(W@Tmy
z8JnOBF;Xv~3nd~oCJOvU?JpD9GP8H^>0;Z%vK<8qgch~jPz!3e)9uw=FKD}|Ak89m
z1}f_M$>PMsakMr}sX7i#3eAvg+A@S>h^?nEe!Q>9iJZ1+SgTteIXCGq<V7~<O@97F
zBCcOrwR(+_D&(AP1t$=89ms>m&U3hg8Y+dYawlN-y*G=}R?E#+1ojvmqFJlM&oTb}
z&PQ()gFc$vpjKZ(?PMm54`wztFR$-T&P0<VXU?RuqXAye3w}~r@Ftv32L?teNoglV
z{&{a};0Pd5wMPq}#gnuXa;}YvEHiI6U*%<hr$}HG1lZ=no`~_gCrHwvo3N<<1W^nh
zK7gqeN1mdAc!0e$ROi<{BiPo7FK)reDJm*T44x(2S!BMuf}~I?I+QeBLCk#<C&Ni<
z+C$5(r|P<96AcWyi}e~gvON(JzJI-KJy#!0qbg=9&LOZRVN@$=pC_hURNHOS@W@it
zOZRJx{}Qbju<MO0?rhgxTy2dxo7_js{xIh@pJ*V&9j2W*DFbnhgLQPy5RIO)4=@i7
zb<Q}E>>EHRU`jKS@Y0fZ@7C78%E{@CG#n$CfEPI6=GM#)uSSmd`0~5~^aDvjr&~i%
zziY?zu4Z)k`gg?K0w5HnRS_8_j{+hMT{^dG47o&{k4?hga4CXoa&T}k%S;3ChF1AV
zylHrBK;jkC_=B(Czn@MBLU5cp+vgRj8X`l$&^TO&0U$Hjw|a(uDMy;a_J->eQ3kR`
z*RSCc%;!MTe83nn_0~1)LoY@9Q5h$VYl2!t#9V3d9jxLiPSHoU=}yW!^TX-{>rV-{
zHEoDq-SJ+4@_`&_XfJJt$L6)@a6f+jtd^0L^6a{idcOimDE*sGTV^%^t!a@s<n;|Y
zkuup~c93bqhs`QpGxzM<C&UoPyEnfDb8dY7-Qo!2Jm11%`NH~(jgPBw4NlR6!<&Sm
zk`4YeX2%ZfoJBf4j`klqR7ASeYz0zD`#ZxoGdooRA7M*>X4kOqyh?=KJ2^7Uur-#s
zQ5q5<7C16*hx!8Ph8!4IcifqnPVegcLrDv^CEQ(CD>8Lb3kin2e4W*i+NvD=?E#a@
zO#bm)yl}we1oueNaRvu@B*MxO56l*B)5hi;L6M*vShHBujP)BRCk{gM1CHc!>C#}}
zv5V-agk<}>wuAD@u3DR9h`45hIwWf?+BoY7zN{D4&(?^ud|leERVx`kkWg{v*<_zk
z#LtV~>d4s^2AwyVf&+hJg=(UsKO_dKLe~Dne$3@h(XoCZ?S7XR^XR?b6^TS<q`vN?
zg%haT6CnhL@A1vG3xu5m04sA?V=!B2Sg1xsyafKtM(ctt(Wxoyggw4u92u5U3)R-a
ztk>MdWX!H%jX`nCK!&^}yw4@->(smVEYf$w$SorT9s-g_1~QtWQ(VC$&e1RJ+7<Co
zn+Hd0C8cq>NvU?~i*7U~Yeb(1Z}pE!fGNoAx-{<&I!lkOzc+$G;>_~;zrmURc7gvN
d?|atJ^nkWkmxyAcd{OJ8KV`OVz^^NI|1Y1-!ubFI

literal 70579
zcmce<2UwKnwl+M*s8OTt7!gH<s9*sBqoSe$b{tUYAl*hU7Nkf|Y>A3x6p$t;eFg@l
zcd&tgfFMIJDk1_>q=<sRcP|8!eg2%Yzw5is^>gj)7}~t^zR$DP-PYru%KPPKOcR*K
zU@&Gd_v}<<FeW)L7!wcuFd3iG|L%nW{x$iqqWn($8y}x{q`Bc=Q;+Y_wqY<9U7&v^
zTy6Sb#9;iwVD8+e?%@Bv?(_kkQGM=*;tA5hzfX@ia6b9>IUSpq$VMD_^@D`jnwFrP
zn^OA@n|qbK*{dz>bwr`)rEjp3PW6!#Nu$G8uIaP}U0b?xe(UOkq7upO{ykq_J(+*C
z`Cj9LrJWVFC-l41ygP>m&CUuP>j->cx!!Nm=w&g|4$I1_^AuMkBro5r;c%s4_+!#Q
z*P~f|A997-Dzb)cuMfq>#`b1UoVAA~VVJpU)~s2~(kqKLuREB{a2fvGEU@WRgs|C*
zuzeT*_~W^^p=+e!aJTm^1F?9U=KX!04Tgsf9jZ>W@{|4i=I33zcUMO}s9&fg`#FUE
zn_22BWzqZiXQcxJot5`@T$TBF>+<g4qT4bS+C5zTw6dgQhS53+3i$S;k*!~P0!P%t
z_XoUbYz)qD>TkVsL69ZvQ)rwqGCb(J#^_?IeJ2;6&p){OW^d(hCv75I$KKz|`fJ|a
z;Y|hJiY(VYjv-%7(uu|#^9pW&u-V|pcF)u+*RK65B_+kICeiBTi4!V2ckcAbsff~6
z`|Y>i1f)-f_-;7Hy}neQZB*bTu==3;kCV^XwO1QWwJ+ct$-ROv-rgT5D=s7?)YtMt
z{`&RnjxT?kvd$=f&gV~`Oxwz%&!#1}{opWR^iKc69LSX$zVrL<-_d=sZ7UbQ@3wdE
zWnr`W^wXZHGdFD5Ak$Z-#tONAB$rLEAz;YS6c<^(+@dMpv-`t`l2qHaQ|;AK`);k?
zhWns;RZvFMDB#y$e^p>IOOj7Co_b&TYqWlvW@culubf-v6(KnZNy!+6$cb}uvwxcL
z^;z%YSj(?t9G7_ia_r_XmhZ}*{`G5&{jHg^9C~_sJOcvsc?DyibXU*{X7e9UvZ_zl
z=4eLK6U&zx&hc1KoqQtL?tSGQo93dth6ck&=Vz#SOXy#8nSS_m?$`Ir*lJc+G2xhj
z!EHS38z--3IefT#`0(LVuO3d4vaD1d9v<H0*c)+Wq4eEMm%-L2{D$!h%DwCtj=k{B
zCocS5zJ<p6DJOU3UKOe+4OO)5Yt`UIX~*f=6yExE2fu{g^GlMM5w1f6%v3&sVti`q
z)TyoY8EL<tIU{EEx8Gwwxs$EIS?AbmI`rYHoBHnEyHlD!w}c1Y^zsVBI@8&%!(1ca
z;n?J<CCt1uSDaO4$-Q}E>@t)!`|D=bPnbA40*{*&;<w>gYhQ(-WW^S&og2b4=C64i
zR`}@Aj)2Yf<qytGvHkd7`?QmjmczGK=eZ=Esi>*Z#cG=N<B#h2sREheJ3B52Gx414
zYBt-p3NtsI?milNdE(bsbdOE#wCmFJQVZYDOigGjzm0pX$ik-3UD;@zWLdSw<#Us_
zT@CgWr?Rs0u;$FKKNghw%S-Wxp^DzZOeyn9x1Ajw9hyU;n>y@Fif?P^==jzpa<pR1
zTS`L(WSx^+D`E_vEgE}=$qhesd8wT?WNDUob6iG-`gO8gGaUOntfo2~L1nljU9hwx
zZ~w^9(e`7Dbe%t`7D%Sp=iX246y=Vzf4DR=x-QyQrXYqhP^_mR+u!=dgsae5S7K-)
z%!!fu(yu9%@M5O_2~(>PiCNk%2A=)HysCZpnJks$rV5QA6YeG#!^L%*`!l6GWG;9K
z&RNTA?P(nt*&r`xpEppY)szwF)6LppA=m3&r5UKz*MExPo-d!Rdv=p;>$~ozg4EMJ
zVYRijmenx^8KQX~SDf#@l2#YOys_@Y>z#gWN?Sghohi8GK-eIoL%z;+p#1(M-I3D!
zeGGLcIbrTeQ<)J7HN}oMzf{P{wpRZB>noA|6W0n&zFVks?Ai_vU(BY4&Fnz8^%n1+
zJI?Z~ofNuNdB#hF-PyX|8yXshhK04Z@;Gd*nA`gXi{(aEG0n>&R6;MO1!NpN!1{6W
zj&t^mh8fN~X3E$2w^r&;IWy3(dBQREBrlDO?>F8G&dO164%pTAq~ZXBx6S`V<{2>w
z@s#O|pzS|6Tza1A;zYl`x8;T1nd-7Kb*5_1+d?rTy<LwJHfx&McnXd<YS(@6e#yR`
zVRZe-^qktwX#=Cr(p~(!`X#;ZN_GgjeYv0crCV{*<jFb-b;&1oG&uKow3bJ6u$Hv)
zlP^}L_o#;!YUrG)-jpuvZhpm0ow4vjoj&9E$iNbl*d1*$t<TLF$}&H7DGa>2ASBG(
z=eJ&?&{s-?DXtT*T;L_1fZBfU+_{y%yez-^6nj+MApO9y8_bfDk~Q;w)U@%mU-;58
zRb=zW1v?VnN{;XrrFo+wh)T%WUpgZ;*{k-+)QtmA;`mMsDq2iXpY?q@D{QaN5~gE+
zoxS*PQ4bD2$2;$PI9>8xleeL!vhp=S=dNuoJ%t<cpFLB@e-3<Xw^*x%BC|id(4B2C
z!F|xb(f+!^d*SH%g~M09XOFl~WiZ^QEqy6|-rL*TYnAq0CAlGWL${%XkL@+f-Q3)E
z?b@|+<HqEt`pHd;rZ5=8m-wG)EHVh9KECnl;HH;S^M0(pcWB3-Ll@iB((J4}7H!tV
z-WOME@4hmT@zUVoiB%E%7rID<KCZfnABjlDkD!|JJwCCDp2BgZr7tImx(gkebUbTw
z-22TJ%AMXxX4#p_jDFGX`*a2)XiwzTElpTe((wWZqfh4!U{T<!LC?N+3MMU1hkaO_
ziwu6o=ll7;Ghj4z=YQTYGkSyKm&*0@n;6*v`0kqpQO*YG->Xmg@^lk@vbJXQll;qt
zPMK!NFVgv1_Oqq`)%D#9e!IR2yfeYY!cALL8N7X8*CK=Q@(ulnqU{WqPm1%DUAh`J
zGvwcSZK=Xuk+ijxVf1@+bscy4@PwZg5-fiebyQO${jO8Z`56n=9e(_SRZZf)p@BY5
zYf~*BYMa`gJ%7j!yj$wAP&$_Ftrq=L*LPdzm!&y`in2Ag?PHvkz%?f6rJiD<jwU*1
zn)5_6h6ma-lJ7bN<{edab<NznOzm8=f|8OJmKzTnS>b?oQ}mrK!Rgbcu`TtJ&$;t6
z^0P}edxtui^sm9qcQ$;p@uScUhnNZO-79{yKXc}c=dD{2$;lgsUd?j5_TXT|YtOp0
zO{d-keeP)%`{kFQ?LD{lE=w@yDb%IeSD@{Pn%Nd5O`dZVzZW0uJ+2*dr)d3n>Nwu_
z8EKohdswnxx^{TX`*pkgm4zk!4Q?^nTQkGvgzmaDcVMW~&5&2HRqB;0Z>)o6XRH(x
zQ*3bQ573D>U5u6X<niM@EX}9`Do?~FD*y5IH5d##f=#RtZv#grYtBzUEoNT4dR4O9
zxoWLRk?GrKKD5mD@4vEazkdX_Yr%^bYFK@|(HfKg^r^DC`pxw8^fygSp}+~EVq$IF
zLff?X_;nsaat5-pvM25BwNMP>kH6a9)YMefTX*l!8RH`Vl`ZAd-!56YR3pYfM#`*o
zCw^V}{(wzJYmKp;(fq1YY)n#`Poo8@#~JTqmJSaM8m$xHn>*K7&7%0WjDVE+P2Y89
zcXblXsoCFHwl6k~TN7x4Cx$Byc6Vo_q@{7tYINd^J)Rlc_*5JRQW<Jda#IB|)Ck@^
z!?v}IDdf_7ou|i&H!c*I?~$|Qs^bsSr{5?sEi3ff5Y072pYP1*zfBK|UM4qJ9`!^)
zQ88|_lI^3f07iMnE?Wz2o+9Jr_W+q$dsw?K&s)qiuT8$c$?koG+i*|drsFT&fs>5a
zE?c^E$Nl^FaXTvfj%7w_M(eZy&{lYf>&72`GL6<V?s~-^KP@c_RpbXSHOF=8+&_G7
zy>)HGvuAsUUh`)~^rlRs%GQ{B)ms>)l#Ru%prBA{Q!Ep1Uw>LYFE5X6n3Wkik}KpE
z1(3S4y*fc{=gzZRWMp_5PWFproezBY@Ifc}_?^swR<U#E&+Ft=I_2qTH&2~4ODlM{
z$2!|qb$|u2_3H_<Y4&%2=_wYIl{Ju%kZ3IlzJL}N(wRB99{@{3PtRXfRkateyyi=N
zyhX+SwZ_k$N2-OdGkyMxk&%&EV~)qg%a_Y&fy&{ohZ^iqP`DVT6ew{cFXX4CyWfQe
zx|R;~wej)sg}M#389KH_C#`b9p1SDrSr7QEyQe3bJA{{?_QMa)Q*4?E?ME19xnd($
z#2OXE19h}l#mPT-@W8H_Bk{FL1qrlzy|3g?kL**;EuW}-Xt~nuvJZ{9d~@aqA3Jue
zpl|BW%ibkr4Mo!iN}F)(&YwRoZ|y0Tt9X8Ahe+KtnZf?f82vOmv&wsibdszC&i^=n
z36o&!KyOQMNQkh=vSqlpuJ%S;G(cL~_4W0*_o{%Pt9ph$7rT{P#AJk_GU2*%-@2Zk
zxrjY!`drg5pFW@it??_YjyJn8d6r<5ypW5!+{l+WG(X+fHsbAD7B5~rZRX7KvPd;v
z`e~be{u@ioZ)5j7$<5`s41UB*u+q}fLfc1%J4c9_yo*k%FU|K9vnpS0{A`}bf(`rQ
zOp3iO6+{blDJdy=XJus>Kf86O!oX>twY9YfcN?qzbn_yrX^xE^8&*ly4+3VsX>Jbp
zS!>dI)$L0#z>0#i#IDnQJT2RfTIuKm_qVI4sCc!Fe9Dz83s(}78R~rjRBW2koF#O8
z=NKfkbpG-omlcwCBYI@piY;68N~68Gb)SQKlNPKw`iIChPhoz3{sNU*JI?=j|NB_h
zo7Pq~mWBc=HZHESTJJ@;lDBZQZc@?gj=4qr3l?yw>izm_NypV;FbJJu*}*80$Lr?(
zNc8P}Rh(yFU<{}Z-{(~wIa@%BE?&7(enH69l&?b}c8i!(?z%mySmx0W^_k{Xv7z+X
zu*Af)xHrvblt%OC*9mqgsKyx=QeDFrfH}P!C7i%!M3|1R|6Cr$y>RxdDfrAq503=t
zX?If>j}Z}fzPPF=tULwQ<e3Xid2id>qrovnB_)-ZOy;g46$gXP_G$LX)fQq~<~}eu
zpf%sdXYVq+tNjTU6`Cke?c9{r&-U-xbI(l7`9W;}3g_~ctlu1%q87JN=FEV5TT4SP
z@oe!bAc+d9*X0-Ql$X~%mv(NNy}7r@YL|hQ@aTti+mnQBRDRmFZ5zwFePQvlXYu9>
zQ2FlOz5BjAX8*>!W4uN)b7a`5phF8Q!0#jI9JBP{x#>JSy)>J7sB6uV+;;0i+t1xb
zxJFzb_vzMqV0)|()bXl#sblrE9rGk5C7Gol42lU?@_pq~`v&hFYmjZPOVheAbCIgK
zIa^gttvz#9huyh;?fDl|1M|DV#5KVBN?9U`@kjG!^I{AQCr+8Ee&E2ZhYue{U6~ch
z{kkm~jDtp>yPH_n8E2p6=H`YCy)E<`XywH8<}D92#cg0~$AtxZAF%#t$E~l@tS$N3
z9vcR{_x{=XCXMpbSToqcA-%n)63kxgesp1$7GB}Y&=6?XS}E(=<VqdS&^67mNjAy2
ztqObgDDU2Vo>*XeO_BvFs8>J{cndGBGgH^h;`nX?JM?jH#flZrAuMcQmR1<LMNgg~
zP}yE_XU7yOk!g1As(6j1-UlSmBd!b6=Zxd3tXKQozAeE-Wr{N^c>46-;3^4&bknN&
zdIFwO#qUajJ#XLU;zip^<b@K<Ii{6*8m^ySU071%?Bc?=Xi@aLcc#z=Lc_x?P+%O<
zS9(7+Ugbaiez$G*{PBzW)RESC)6b6+6cn`Nk$y{-N#5M^5BfJ@GtB^Lz}nv>cxLsn
z`YlY=hD;Z$>er2p?N_u@PKFA%kSr2et}?&pR-th_lnaop^{=?6+BRNvQaKcH+t1J9
zh221}S%9=nPHQ!k2VSuJEDu4MYUA6~hdegaUsHEpt+xoMZ1nbIA1=AOi*?a;{((EF
zXv|WqUR&&uv`lWk^1L5rF4|Nkx_6ylq3>FgD~VP$VI6g8Vyjm_H-7*6$tCf{mUqBC
z%^#kfdeYXG7gsE+%a*!na=Wb}M%Tr9^}IK2ZJgR#y&=4NO{?bOz@#trC`aDr`$A7N
z3*-lPrPrd9<(hkcz&f2XfBs4>EiKNlpr9aCeqg<uw{Ds8uzRJRJbA)oZL;ssdzB!)
z7I1X*V-KslyL(l0=Wwf5mT2NJG$HZasg@0yL)@6?HIO4zz*}Clya;cCWFzY`uwu;`
z)mdA*+zIVNFtMsj<#l|RQM{$+vTKiDLxgsmG3hPn{h>e~1xMGI;m4nJ*hzG5K}Te(
zg<qK|BpcN*)M4+{r9aSF)B@B2DBr@<GUR3dHjx*!ZQ`UUQ#gknov+Sv%L?1)7v}Hp
ze^M^)b-A)8s=j&2-E*^N&nEdO@5b`BDw9CK5|6T2BSmTg>@4sOGz(#7Yh79r7QB>s
z*@_wSR&VWV<-LGLFz4r=k8HRWvwwV({gamSN}}Om@C^TL&~(tO*ICzYT(sG4FZRD@
zOiaw%;=mXHi2{6dB=^@lDLM6b@RnZPa<{Loq74ty5e2BP@v2<w{aisY+o9N^IHQ8O
zpFe*#!_ufuu<&jg2-<e$=FOWSkULuNm|y%pY5JR%mdH16jshwxQlU*sO2QfyOMjFa
zA$8N!&(DwU7nHd?9#7-Mi4&U~x`O6P8btAlX|C9`Nynl*>V9?JVC9jkErXp|TBlCM
za}6@sJdT!_h{!Ht&a;31IiR(;zV4x7`Paw&3$wK$D+}l@a!?L!imo`CJ89A+K56+q
zd-hznx^w5wdMi)qiuUg&PE4?@id|B3sT2zz&pF|->#J9<o|H?rZz+uq;Fy`2+1331
z`|lV1!_njO^Ydd{3~N(qf8GAc^4JWK!h};Ts^a>oWp7h2T(|)A8j1E<9S;arUFR`K
zAShsZ(!DdIkR6xiKWBpE*DGXfTsgir8OK$}tIyV-u;T6l@#<lbXD)v`q2uxRTXJ6l
zMNI=ZB*TP0J#%bf<A=X0WbWqgF8SxpjP8%v!I{8NcKNzvw@Td%OndTxZ{i}oSq%5q
z-<ArDy?izkgy-v@-ES-zYi$h1@+YG;XR8!HFWN{$lDbIK&N10d`LcC%n?~PkkiyaH
z-6~Zfemb`CQldfYCtrMH|Kg&*{hpZy8xuA*ei4nxw<czYkJ<&}OY&me$0Uu6-%A%1
zg$W0xzWReyX!`i;MlfE6eC=To9o0WZwDRGLLH?sZ{=U+AS@xa2$wWxoIXQweW2zcn
z@!>){l*X<~>)xGnjh8l+3|_=fbr_9bYx<EPwgJ86{69zOO~hN*J$g-DzQg)qo6?to
zr4Bs}Wsg57#tNZ-cs#hO@wdSE>jnnxA;tv;Se9s|%VTRO&RDRXx}BG#A$P$BYxTbF
zi|c^M=6mdJdiz$hPnG=)jP&R_0Z0=?Y=dbXOU7TR64K~-FdrUdmBN6{TFdwQD}%pK
z;islmn|#9Y-JJ`kUO)a*frFOAbsHIM@LZz>W`o`%X2i|c^y$}L2rTpEi?h$(ur!0|
z8W+ZyoaVG?(|X?pUp2=ToHk<y4~*;-G|GUDC$?8`_0^1x!wei=O+tO%k0r+Q6w?yb
zdvSD4BsjFdddmYMwGT21uJ>1=5f%o@8BUlm!9HUyd*=9Jj(VJ%Yti3PC$Pog9ubJo
zgM$SJS0AVra(FqxrR%{I@G;iOlP7!8#WY`;tK6I`uJf)xYb3+j*_rLQGWpW@O+9#G
zXuvk()9XtoZERjpiR1Ua+Zarcm9-89{mKLe`-kz@TtzPg0PrrQ`L<fXCfkecH7Kj3
zK?7p}R<CbI52=7Y!}qU$5lf5Qabcz-BtueWIj%A3-51TvqgFDh07J<KGO>KCsHphj
z(}za%fTi~K_9&-{)A=RT4GawKw{3C$bWhOj%L!hzuGTEsPeHd&znzbk&0ppAkCd|%
zsP^K*8!Xn{UytNOs08mO^^K5!OxAGfvwrPrlV!~L^XL0Y+k^qe*f!?OKxJ0}^3phY
zGFI3OM0dTUwk`-NE)7zUNxv*E5KG3k=k30Ng#X08Ewz>YEA{3)H94QJ0(l&pY^F+`
zjwZG-J~n0px(YJ^$5(2M;Du^3hTD@G*!G?E`2_`PxC}l40X8wa8_Si-FVA0l>fQ6j
zQWoWVS#tyhxw}0C%v#GLyFY%^w6u%@nu&g^o9=Kw=c<q-lLTbA6rSh;&9|NQcc%AT
zVcoc)k>%>*XS?day{(cNy)1!sra^)-j_<l&J<^Fi`X|X>*f}efEP2Qt0sV^k^vNM5
zIXOc9svutFM&@A0X4({J59A^|nk8o`_4dpxS!MJvqi2$?AEsMX$1?#YDyB=O2hlT9
z@7MM+Sytz$t&7WpN*s*_vvTE1CTrDiNg3WzSUYnDUniueZ}IbfVRZE93Xof%BS*-0
z*u-Ko!=D<fV~s9B6S*Uq`Dr`&vB0l~&!(Pgx%>3#9(pY?&8X+VKb}55k%nJh^V_xE
zd0qKnOkcS)G}y1mg6Cq;Aq_MbQJ>*dP*<l<g#zp{-nLEC$i$@Wg_0ca+;o0i$kC~@
zcWmFj4(hsM9lk*)<LoLaDGj3YXqE6sxHoUkApuKTTH0~2J<0R>b;4h17N=W6y?Y^i
zwt^20bQehW1&(}?U$EZtn(&N8o7AhSs=PAM!k`~$LGrR~DHXv7ffaPJQZQ{%6pN&K
z#{%==V_8F*FlfvhGBY6HbBNap3%`StK@{}M=bi*8CbK)^qKe<!k#_1=NQMY7RT-Gg
z$)QZoqo_*s6tRfH@!$x>QJI0zA%urY&>|r|V9ROM49C7wyk(#@2d(lmx_|9_P)w#O
zJhXfnc}NKO&GD{7jtN#Z<tOV7t~P!A$4~TN=w71~dYfkgt0}U;l*0j2wYh1@$@KU@
zi!}}(zFS&aYMFzZW}qa~ce6?4#N;#Ee){PrUm2&wy~}RQU3K7K3lIkfxZ4Fd9I{K2
zugSaTcUVVl_C7)t7iRjcw_-t$ZE^&C>_eT>XF-U$IJ@)ylHyfR%?w@oJiq(yyEiE2
zhoid%Kx~_&_axmOQIyO$YL@0wk;LI}!uAJ<GBp=(5*7MZueNO2B3$QKrE^+bJ9a0R
z%f*dNtdTwWW)HJ;LxZ#0Ydy%)u;JA3qP#d0r8Jjm_D_}x3s=U&MSIY33r-+2)pnWH
zoxhPp4a2V_QKww(9gvs8-o0gn(*P)|%S?e~B=l25PWQa!v$>F@{>De9=h<4mX>SzU
ziW*l5M_SjV?n>kn^f}b&Y^*Lg-nrd(C0W;<dYL@|s7VR5`HlUFknnIZC{(lLzT6Yo
zU>yY#iOsTy)pYK#z37ijrQ2JyUw`qIH1vWh3WBsM?RaNds;vdH6dlI0`u(IB)~Yls
z(06m&jG&;cScNaJ7i>{SU>t}_N%1-^)lRbyo6EOFQS`$u+jCN4-Or$_DIfO8Ke$ey
zfOkIa(s&D-3)H4_@8EB;BVTO#unHaz)AP0KtdE95%EA5ds`CXeD?y#QxTvkUS*z7H
zC3()!nRn-`HXN&uHxm<`vwoC7UdID!(y_Bk1hBh_KDyqj+VoS`t7(w<ppw8g48yk8
zEmTof{w4DDE8|^(+D3u}+19$vJH`#jOWV#&X0ou9f@t@2Fa7jUFVk5&&A!v?RGY&@
zCR-)=Jft<M>(Eu_%)jTQ`D$h*C~AS_`kDYq85w=nP*)FnljED~@|N7@H*cPz*P_sQ
zSwOhfbaYHjJl4QP11w{^9lGU|2E3raLjAm?Hv&Hn3(we=+Yul-=%eu%k44!iK}FVo
zyz)$k>rJXoesQtZ(W4<epE!F_=g(C)lw{4M)3jK`U*5qE+H~^GA9W`Ryd|{`9t@hf
zaHH~TM?F2gbq-wz0a99DJ~(4HIOp~Wi(|PTAQ+61Ni+CHP#g5>?r+?@kUO3QIC7!3
zuFmSAev9TUmv9K0Y*;V@Z~cy~k_=DN2d)4WBWTZe!hS@<r#NckxmPEwt-1SZu!C)(
z$WAMql?xk!x)H<O6P#Ng8{h@V<DA*E?~($jDVG{}<3{kUTl1*=?+#W$2heOZbp33^
zBsPwHZq4(8l8*hXpYBl7QFZ^LLC43O1;_TV&~eTAw!-x(^jRy9cBBT3!Z&Z8>Du)Y
z68i&P2O@b1;5{prE!&Nz7N2fecyRU8iV7|8=>aIP(F1{UZU$fjECN)VL`ar~taV49
z{tG_ulan$Kd{7BlA^ZjocODN$JzXm{9cm)8baQ9=kp!{N*pJXT<2S{ASM=NX3f6?C
zM#Sj(b7j=w7SwGou!dB-b~d(bKIC&D=dRzdaQcefhF_di-?kXsGuklAz{kf22oG{a
z06D`zR~&R`(u+a&)Pr-}lxEo?4ge5z>QmPtu+FR(cXwS{Wrg=vmVUbD6!<+n{s^T&
zSyEmn&e{^RQZ4K$RBPQN>jdfQAmc(`^y{*x*O!ve2t!0wTsN`IeZs_}tLH(%mEZ*A
zx((Je2s45EE?&G?%8AKj|2A<78=C+{OcQ>3g4>8I9y_%sRtUb-N`4qBG35Vs$6s!p
zAtbAJ`}S=LS3sbeGk31$YZtF(A5W06Kj1&yIeYed3L$jA$#eWzWx|V4Sqf67`t$Pr
zQO~~|-|`I}cXf3=FACV4;M8yJF@G)Khg2!H!>OixKAC}c!6b>VS-ZB6NY^4cEM)5G
zc(eoqvEZAaH6}H<>V_)?YNE_q<$(6>a&>jZ>tCk`18?rnKL?}Bl5m=PWzh>fza}UW
zFk{+)3DJ*YePYmS6!z~ghsa7g0pMB-zRP#@q2I}$upNf1^5qpeCwQ__pvy-e)Q4Xn
zyb&fXkJ8r3a7;+g$lwzaI+mbok57<D3=N*eV#R|%6WoN=R1T|H(Iv3yw6PI3Vi;<M
z1q5OUCM2xGA&z)*iQnSMMLrTcQK>5P+IBkRjgh+d(8s1tovH$go07A^s#+b5KF<HT
zE(;Y`Ok6zr>BsEst$4Ii>NYNrX}t`u3P|OjEVpK&4iY3uvw}aSrm3lUy|J9DugeyA
zQfH!lTiet$f>$!%e%h>ANi`j>Uflu1HOIzXZ-t=1dU^ftL94Zw3k!!lO=Oi{$bax}
z%=DaE4d~h{sVUOwOGdXi<Dk&yv5?8%C=}x#I4Hy2T!gJ#Fv{6H&wt%Q%F#!UX}^I1
zuC_>b7DG9DOqODM2JQZ7)R`2WVe%KPJv?QUYzOHuSBGtiuRlI%W4s_E$nntEPvden
zuC&8<D}PrXKk7avYY`__{@pt7#v<K&qh<kvvGv3Go}u~fPW(uteEs;hYyDd$<>Fer
zu@y|e2^twhGu1<HDzP0Wg;sp5pij))jpng+YSF5}qC0(p1JfIBj!NbV#L(Plhsbu{
zgqi&YW;T78eEsHbo1~4&hl9?vjjFl+e3IH@GW)+}VJ?oBm;Wd){4;<3+s$|PL@mfO
zUFX%>ILqzR4h>2Eh;noFzA>TT_^~{&0?<uheOPXn+S^)Mv{y>Kc=195P8__t5Y@L6
z+0(!6S6ko+#Amv=6~FrIHRrKCDfk48R@dj4J#_YemLj%>!qdD=f(k6IrsiggB7bRE
zKG$G*IDUK|iF*Cia=q?AR(HM_>BPcjBf~>52hqdhe0zXbePvxz0n*fwO9I%~)|oYu
z3DL-0O%6~7{7D_%9~v=PJ!~K$?CQi;Z)9v>3l*R+f(M{vc+Zv4i$Dom(kTw1=}69H
z5IGNZ*ph^;;nE{-wn!y--4G&JOM^=ey!pV71Q{-Zq8|Hdz@dBVojQqkLZ0*j*PR2k
zH$FaoNyYcM<BH4H&^#EW;3sIoLD;4iAKzEe7RI|tfXoiJuN4~Xy<Rl+3N)6bOt7tT
zr_PLUlR%g9QX*@u52gr}0rK308-l(vG}tgw{wD7R6)1GLJhm&)#Ni`HY!SJlx`gJL
z>ijuYBO=w|!(kH6FY!xiV%zhf+oJl(^w-)@tAVRq=t9(;(r21ny_Irua&UpP0gY8r
z50>q{b?`;QPf`jsSpap={?_>HwR0<;X2`Y=4|E&WhD};gAej~Q{=Fr=D4{Z#b7f)s
z7Qij54w!Z2+qOyDzGDY8=c|Y1N?*Kazp`kp@nvM9Y|*mF&jgcsjz<Ka$b71W$Vy@1
zogkuK!Z0TE&>hO|9eO0p<d-y1@2t;g>geDumA~@t@t;3er`mErtNUPRwF13B`G{Sx
zf496m&{rGuc%rsIxx&oI$VdS>*K`zR+b^F^gocEe0wd(-<(1SN%Zk2y`Nwk?F6b8a
z;U0^kjR0^~RK=MPqeZ*YM=$^$BiVp>Vn<p!{}3>0Hx!J?9AKkEg_Cf(cC8xK^O9hB
zYFNxtkUjl7pH-y4Q64RrCm#{7d!Gxj@BI}7;u4LEZ;SJLkK`y(1HQI+$I&%DXm6gg
z+j5^gSqZ=adaH$N6t9>u`3wx@J474+4yrR<(h4i1qE@-UtfO55e$xbd@`cDKqaY`;
zLSIjB&R>>V>kq#(Z!BZ70DXmF@Nk^IVBJVCsfX^cxS8N~`h9hRB5O48s6T7!)DpBj
zz6A>`Y-=a8Cyz6l_aKo7ph<jO3W>j?68MXAJn0(vs<5Z%JUwthJ$-$?g$pf{T;4pr
z&I&n}`8g-7-N)p%3Rqqg_rI88McT=J8y7F$#-iLH7D?-Mc|jG}G*$z6R62<3l>*5u
z)C`UqDiB+-Vh?NGp@-ia85`>)00k((mqhF1WB+!Mqw-h6A<tl|t`6uaSGoNgX%Vrs
z7u}-v5gTN^)uK(O?xooM&HFkSd%O;<|AsKQxuWXzL(#&#NVVKMQu)<c>B9pRS&$*r
z#+3?K)D|zZe?z@d0|*s~lIQ)kr8)14iXJY?9OokK;36hA?H`ZK?@-q}u3u>f5r>)d
zyD=sCo=8^yCY7HSc6UyKU+O>lR0CkrBFu5CYNf^2&*Q@F%iC&5E1Y(8Yz1Ap_~)OI
zH-j0|>=OjPx>6nf=OjJEe)PZ=TERqplQ>zAN|aYvxDPsl?H-TVkCJ&r9xT%!`QHr=
z7MJ8I!b!oFqN+s+1WCK)zjnxuGsA3|mJL~-JxF=SHmkHU%ydqHdX5_Zn`&Y6>(_T-
z5sGctpfzQdV0=w@S1s1$paP4MTv%i85Ns%cXKFcMVr<NY?&-VU%J-2u7TTew{^UXR
zW3g^R!7uCsCE<Vx3CK9c^K8#N2>ND2o!odiZ7lo(t@-SzZ8QwYLy-k->=ZUIC-L#)
zox9}a$)m6<t`S!_v|1WWl5_Bd<$7u#1V7P%toyo?|9dF^0FS#gt1D{MY&w~K=2V-J
z6|%AbU<s%9{||1H<=VMdo{VQ#?}@Ejr;Z*X@=<Z`UUd(FrmH?6SBL~JBFivqu&xs;
zY57<ThBAy~PE&=25p{J-_aB)<A9aU-W64PZHq{8L9)t1}76ejWfkH?M7rqvX5=+<_
zkXBP(>_-qe$*eE2^)CJTr%#`bbQ>9FqpXk>2irwdR8#>SYG$;G*YLmFOv?kERl$06
zU>SmIhTWENGRF<mEY3T)8k(^z@hk{v;&h3)97sVtGDpFRKyMG6D~k{qPwBerqo_ht
zb^B9j*^S7D35k|f^u`dz+XHbRR!;~D@tQSj;86wlG#6U{^pY1y^%mwo2Z-8;%lDsf
z`24jN<*UG-RxVkx9hN(ZbdG(kq3Aglu#+`ZgX4$lYHNuK5ha1ra`Doo7g(JM=y;T3
z0BfV{PvQRhTtTP34Q?YDSpFmnM;m15Kp-UdpWH-g^Y1h+EgH9ASw2469rxk4L;Kp5
z2)`Qnm-@Gk=sccYUdHx&QLE6F(p4v2giHbdXwJNO5lAO^R(9Iqhjft6iyYj1cemM~
zlY;|gr%3y)vz?ppb3bTcIfORF48e_p6C_`TyuRGCNbkDj+fgLQ=oh$R`t?t6XTG&f
z#{P@IPbNI$tr{_7U)8wm&VO9h&;IsF(~NCai*&})wxQn=)&CLw{g)T`M_3stppo(Q
z1>K!=Y7Djdroj#xfBOIZFBnJVa~eJCV$}O)D_Cg;Y7=6gne#WkFm8kWuUzRr`D_32
z7yku-mx_zGbGb&4f>EXi;9RV#SZ`VROK?d9OcG7gj(p$nD7%i@mG}e%&W-UB9ZFJc
z$FbN&auQN+z)XfnOHq~o7_HLSk)sF)60(eUm1UydTA#i2$GoHL=e3?vSL4c_41Kkw
zaC7YzZ9Kjezci}Vs1m~D88T$u-QArI3kHdf7ys<Yc+-;YStCQ8oj#}C*^kGg0)Kml
z%4QHk*SUUiKz(=$@tLGoz#&g>Z>=};ARGc?m}a|A=<e!LMhvF6+9C#_3q`xZV^y(F
z-Zq(<6#17t^I1!hD>gbQD4_q8*@NSefLt8M#`%(OxH8gPIhW@H7z;!iWH{1@MJCiI
zg^XiP07k(;%#i>Z1c1q@HY*i|O2OXS6EyJ4nC7!J^Z-l)aNJ-pTMjH^g1`8sx1Bnq
z0W4Q8U!H4xxyrH}E(%4+D8qm_J{e)X&2l5!z!IfM8CW5&#?(f_gP<;GM_)&sF5*v`
z5Eup!-va4a2}}Q2`iI?r_UtL0BIHc9(~PDS0Ct;;0)T{<A@)Y$1kaNVAGeHr>A`iB
zfCP{vLKqe>v#d77hW>LAdXsujkmI*RkUq9Na&Yq3h45YRz5Dj<<B^YqWDpd5GO0;>
z0;H~Y9ke1<V#}7=upHd$J|tZt;I|Dx?+HFLWtJ%134*go)*v~<S`&Tx*{xqmfQME_
z)yt~#-Zq$m*Pvfgmg~!xFNdMbv?rnYtB}5rNI%s@z$gGF)#JzS(aMo=`l$XXvy&zW
zAk2eR7|<70Raxtg)khI#ggFlQPC>i;-c`ynT`Z}u8@$rf*H_1Ku>DZWKO#v|&hSdG
z7rOiVy-!(8X}`HrO_&K`E)+Vk<@{zy1-NmPa!7@gA6cP?8d!+o011hax{6Aq88DKt
zlXMcUUdj?BrQ#*IO1|rutav;M^e*kfL6Y7{p#dmG_tEbdOk56tf%5H$R4VYkHE%uo
zJM;Ce$(2vWt4O7%r>96<mEIe9$Wln}nODFgfJ#G&PrGK{EvI9Jaep41Y)1!%m&01J
zWJwglRA|;vwnSg}1_uXIxC$9Q_yIFJ+o7<90anO#KV52G9Ur;R35}L~6m%OU)z6(Z
z|CHI%X`Ns{radqi?_>FIR@;0-nfoW7d0-s~<3f=I>&-ajW2f7QE~^aPc}a6ebpi*m
z2fb<$-PW7(V5q`;^UD=fv13+cNp!mKhyIg*p72c4d`I8*F=g6aK}GKuJ%X5ps(q4*
zfi--5@m{;PZTn4_SiPB^Hm(?m0flHKRA{(ZC2-%6on3|qFgW$Ij+rJ~zX91vd>!U8
zFg~9d$JJevBGnp!9Fvst5a!R!*oE5;(r}2VP2>~NEhm>qOQ$@Q#kysIdH`tY;$fK}
z-%sHjSVDb35jqKclZ)MDP~+WITmF{X#$s!F`xQT0;F(eOqfk1QwW_%sKUi_H5Of^N
zr~(+5Vu(_v&$G>+;S&kYpS#o2KGnd0(xUePDs;qN24twUHYE<Rtq?SS&%w;jqwD<=
zj^xfn=(px<T4{9fFYb50t#I|q1;^4CGpT3+?i@+6UhgMmQnZkaHdqRzjE6!JephGT
z;F)){^vG42Fn|`_YOY}x%~^?V*idm|!{U~q{swOvui2@^)w{1(40efxig3$0euH~@
zlEp3!08xKem8**a<zgs2g%G*xq|TZBOUDHh^#foZ01;aPWL<O+kZyeueid<mK15ti
z^KY)C!73~qt1G`cV9QD&(oTH>{D(qce1d`!&>F33lT~mxPP}s&hl0q)AYB~_QPvzD
zM|W>8QVKP5kgmOV@7}bA{s7F-WieThO6YT1hAtYE)jKnJ#+-!<qksj?p+pc^hkf7C
zy4Kxhyyk5U{R2gZioeH#aHIvzpxSr$_3<FR5ShjXuDbl|ydMc@p=}0@d=4B@8LK%!
z?rjMb1~0%V7r6?d^-wXwJ|txvJt+dA@apu_adk_V=G!#b!zI23iL@-nP|gM7Fd0Bl
z$UXi2F<oVB1^`L^v$JzLujsHl32^jrP-N}exY{EVzlM9(c;G^i8C8bZD!dXE1dxz4
zqY&NPMYR&B#cy>mt5KLxgJdm0UFPG{)`7?Hc0w~R#Mbspx_wqmvpqAWkRAc=3rE#d
zjhVumGERg$R-$|W%w(H=%fbG0yytJt`M(uEHNKkkv#lnp16$EY_M&gLGGO07qOz6a
zENw+jlx!T^aVkSun550okN;7xr0=-+0h3M|8D~}78djvUy&2gLIttwT@nHXXa1O43
zf>dth{UfM@p|b}m5&_bdl$wZbHq;m6CVuiVO|wYo>#%LVUIAmPN=KS|KX@5BRSE20
zC^tJ`V~`p{Mik~`IQ)jr_th64gSrU4M&Z3y@cH(o5UGVOEpkFJo83G%1?bMIwV%*Q
zQ3TB9QB(v2Vu8Y>jUZ>PynXxcq%ppH`EqtM@{dOM+d&W@lY1Gs^lA7dq8r0B=IUE)
zV2nkM-1CncW*lvY6|RcOGdMBkQ|(1iDqB6Z46X|kZ845Zd7OeNsp~@6-X+MN#lZeg
zu&%2%eh+1e-Ux^jpERa(MnBZcT==>=YLb*jog%;W5bj7fqX|H$M`G&DX!UW3)I@-C
zm4>zBCX*ZmSxr=ZT-#MRX&4}M91bay1!`XR5wbuF3^|P6g+Qt*&^UrENp}r#+2#EF
zHO^a>FPCRgI*|Nzpe_|nO)tt;!2OLtDf<S_l~;Cf-22h@;Ng`ggu^gpsdp?&-(X$)
z++W&e4w4?o<y5xc2nY~o4#58H%NiNX(zK!uJsg<fbd??ZY*21^!qMfw{BZW>uHUCz
zo60w9-}yZ&jb8t8UT5GBrtGt$q37i`-;yrg=p${ceO^`FJZ!CItLeol@p}pxN<NW6
zr?*d$V!<!CA?`FXeEfd+BB!N-iGO4mTC^wRw54_o4%DaJ=M6+zxOKv)AQKbV^j=yq
zZj$GX8){bR@$u@YnYa;e_;|Cv-tU)hPpVBy4-E}v9qoT&tZ~Pjms*fuTCxNvw~cOf
zofIZAy@m0NT2ON+($e()eH7#rGgN%CDSJy5CC$Lc?3B8#F*u5$%2?^<#>Q_}@7B_N
zbGr-_tPHru#c)320mI0uflZW{#>MTWSxnqJQ6z>6lJUGoV^s7Y5Sj#dKv2f@N#EGw
z$lk}-<p{k7usZeltKXrIaSE|j6k{76-dU#Twd%UC%*V<jQrxtjqRqC=0ru7YHYZP(
z;@L|<?Q#Lovzu6tfuTB#b?|cF2C{)`^^Kdxe!I2i294|*q@Q6NvD4FQ->$4-mrvg3
zyDlYaP*Yvqv?|sJkqo5f8#H(acvF%iBA#~mVxsum*|X{kl;CS&fyS8A;5U72nAMyr
z@lt1W`nUy~eSNcyjg1HsT-MvR0jgR)Qm<eSQrxx3f?zd<&WaJod<ScILGM`I$@WZ*
z{H`8ys3Hp!eg)<><ix?~1U0g)desU}AgupB#%U#XB~7=iYX8NR001il3tnGyYz(+U
z5nUMDFItwyFxpPWC}5vrvKfugUcX--^BDLK@Er4n6D69$ra3@z<zg8l{a{N2fvXyg
z<IpDxSx6_qk3`Hiin|$uHzc<cQmvLj$`t`r4u0l9#=1f;wqP25g5)+hf6?Zj2HIl4
z@58VjF-54~^y2G#+qiqzu5-v{p)<0xv$F-{W(bsZ&J03=1O7CGkD~r)j*fB`uQThn
ztXFT|gkZwN6y~|WX1gdfy0UmPH3|>`g;86ADb3OE$4HAJtKOk;22I}-Qof#rc^_{b
z-oK#Q3b}sTMuU-rYtX*glpId8kHN@MQ)g$4+m}Wm-ge{B$L7AE5U?p`z-8x#RDcYK
z+Uf;o-3MOd)`;Z6+Xg8rM;-3NWUvJ`1Cbnzx$w=Of4b^HD}GXh>GtW7ka<FVULF&%
zr@O<w;c{jm!2`pCorJ%TTc(^k%`QXIk{C>32${O(jAsFYxEbo{jj0_QG)8Rn`>CLH
z4Z?yZPdS!bz~ENy<L??BTe<D<daBHa$3G|N7^)fGc=hOl1;B~7{*D6&=2^`feIsz^
z%J%Wble4x<L8jHh_Np8@bcgU0$QB-);%57fcpYyrR4|*m6S<3!8>G2DBBT_gNNa|l
zBZU!nWWAoEs6)!vgnNM`&MAz(QC^vXf^C?NQbFgSsWFrqF=PYC+)1znC{O&%O292^
zE!@zc?%<FN|9%eQgpVIT7BCdm($kB=gIxiw4-+-8M=5Qnfw-Vw%)Q;-JsGe1o5Hpr
zmXZ8XaCnEhzBj$Cmy^9P;g750AQ1(c<PV=hMBn+4t|w-nRv_|KR8&L&mPX4cl10He
zqI&-R3pwk3z=z4Opn+<T>K1g*Rz$$}`)|Ah!@=}<(6;J0ldA@&oA@|7<eb5ZE|5*C
zJ$VTRP9(DZ7FdELI~hes^ZQY=0BwaPSc==ig?bN|Xx34i;=9?N3m+Xz!YrW~i>)ld
zLQ6wkJpzM-P;>oAn<STrGMzL|f#okYU@h72^r&<C@^Sn{{z@6<^wI(W-OMY5mo3YO
z1`Dl>9TD+dXylXp=-3|kC7m7+iL#EQ8Zw%m^j07PAW=`%I!;?2j|B*r%qxnHgKOf(
zT%({0lmrB73#w6pDbq)p5yYm+aH7Bj`T?R2>VT(sCl<~fIP5_i0erG+fG*-O%t;#z
z;oO_TLrs!y_jL?aRqvu&=VLw|DV}l=8wv(Je7K!lC_2Ib*ij=O!B(nafveEQ@vMy>
z)|-WY65oN_h|xmdB=W;^(AM?==q7xA+QH#K;pjX<o!vxWL;}+O0V7VD_i-~R`xB4!
zcz}$PCcr)`1W{3pV8Uy|&^E`32C2rfcRW0SwBm$Ak0vvTgcwT+0;*$%EFv~m6~$`*
z{{5_w`D=|M?%v&wP$|X(J&ZDL+_)j=(yNkzff@=Z0Q6u6|3KPc^~wM87I4cK$5mPO
zHzF|A?h8~BtCK9ogf~K<fMjaU31~&2yNWCgzq5uAwBx7Qb3P3Y>UJz<LODTRz^fy3
z(7MjHLZ4Tjgn3omeE1FA4=GERED>hX)E4e})LVq+r9jse@!LlS(=I6{B+k&l4AlxI
z)E_mE|L>()o@0B@78JC$=i-g(CR&yN@er3oe<SY2R*Dga{KMweqcpQZ20PM<I|QI3
zDAqTFvGGtMq?|rllHve1Pj)U&bfDpgsa~Ahq|LDiy!rKWrX9SB$P1_WQ?%Gneb?aD
zaVl+*!{hPZI-ueZIl;C?&#^$uBeexZhRlx|tE8%o<`|`d@(^4i3&>a^tIO~MNEUbk
z&8NXdq=POnbyh}kOQs{HIr}J^G|<^{2z(9QIWlK(@{}otcH&x8GAJvgEOsa6SaRlq
zOxx2gkyJHmD`5~IOtXc47X|TndK#;p64*Ms-`A$R=<tOU^Ay+0*5b-wo2a4DH9a8;
zM%cLCxo008;(?L%K%l~35Ge{ktO>X{VP>#Jo1^owh=zCL3;F`;AtraPUAQ2ha7~p3
zh-9<x<A?5URj3{$kzp!R!%uR>3S}aym>Nmz-w4x&5?+gUoIl{PSz0<0nz!%96L*DM
zFzDiyf_YOt)RpQCr=&usB4{Rb#tNJvBbnm9fqLbvZ*nFooPQax3*?&?ND`Vh*ftkE
zGCob04K8uuIqeIS)o1lSU3&E$g05a4E?%4lk5GXH)zPSb0RK+}Akew{Pn`b}^_Co1
z?1w!pjM|#c7edn&UI_=6oT=1h2yp)S=bup8BJa!H7q94X+A;J^Rlfjdy<eD#vPhDE
z0U~Au^$_vnpa>y^5Dwgf5#AG<Tn2jiI$v<KBx&MaTU#5<7vstqbkGc$F?a~{3$Wha
z8+4-mQ;dxjg8Cu~;|4)unwh06B{bd@C|<ZHUKU{B@y?%5qEoHdxbXn(Kls0tfg3AJ
z@bQ$-O!jUN?{joH7Li4dEW&g1$k7owcUmGZRVm?&@<%&fC-qb$U=g_@5Ub$;*|k>1
zh0~BJji|yf#fH#IufXo+<L3{jSTs@*c($z&ps063$kp#2AQuQ)FBL5+AE0gd7^tSB
zZ+2f<7zuWvo)&hjArmro1vERFycO3=Rwh>*DjS}O2+nj7bM$KaoN>o>TvKxQMAoPv
zGYrWyEHx?apTIu|C={lk|6ms3ZuNj~hY#8Ut~1RELD@I6H54%$y}loyWH<<s5WVgJ
z!?ZST+D#z2H*el3a$TSVv)e5!_NIR6oRK9-@{-Qy*Lz;Hr5g5Cd}U?s4R^mqe#+Fo
znGDKG<ExLcRp#3oVD$Ypqx1j%%~BC<V<_1A=&bdBMz_soWJ~NHXXO9w%k_?l;t>x%
zG#f^=%{Ff@N($rIPGxY{{^uY5|K*L1-?HqxAt5^vtcJm1tAB9^!s1U4u8x2*!d1M#
zM^W(xX4H@cjmm-ix*RSpVlM)@U4!GJL<}+DWGpya{oPWiQ<S!h$;$wszP!6pDy1Uf
z@ERXXI~c9;k&w1NChqDso|!oQ2)+qT^P;?JRfKNb3rq&}2>eT@A~Z<QOu1rACSY6}
zI}EAZc7BA(w}$<<hs*sF!6GO=ZC5sTP+np?;`f-+fwo`PA%z?xO^jmrrY!ly5tO2q
zwwSC$ogQ$-NMLVP2oa!zs}V2EF)zdjpp^5c<CDJs{wd%Ck|8DV7nC^$4}l0&!PQBA
z1MO*r*pFe7dRw<5jIPx<=jm9HzJPm()^=Z7=;xo8;znWO<bX#YAOYeMVGkcYLaz(M
z^dPDoDRElrQ@E1ja-m)VVMchmG^GLs3ofPxU<{@t<-yXx4~`{R5DWx{Gk@^=B#LUH
zY$u#-Qh?98kNUEaK9WbnIA~}O3+_xG*l=rP`0dzd#lTv4hsY?#y@81l5wW7_&6{_K
z(`dp|z#&prNgd>%BN|Hz5z7B6%n^V+a!~xC)<3azDdM=C&;cFvsL)bZS9iK!i$O+|
zw$THl?%h*{hEGZ)`6<YX<NNeg#FJgWJ!K8VVq#DNkc-(5G?8CaR1UU?!KbMN$H1~;
zGCun9q{vl)bO?2SG;~B55?(q;5wL3{!c54AD6kMX?X9&bX0lLtC_P(HP!7~a4~zsT
zaBB+uKq8`)>^_9qfw5`gTX&GgNUq8b?2*GvKAliQiFo|(8^Tx)k@xO(rOKG-JK%sL
zVB-*QMa<@IrBkO6;z~HtxSLr@hl_3E%t#nl65KhenwsP#xb&9Dld$1?>C&YeA?EGV
zQF50u$$LkxPP`E~@oVM-T65N$`k$BWy(SD<bT9sizr1<#W(%AgK=cc(?tS8dc{n3!
zxZiC=i%_8O^5x6va4&e#Hy~I;ta972B|O*?wh^lr<CEGR^1F7KVJ#4gz;xsy_4C!^
zl{s6Zg{(##kpcRx?crBwOy_8^qN*qRV<-$gE)@U^7|&iK4TpV2tLx~nQGW&crF{14
zBwG3)18xgWRi&q9+ij0L+4^R;x+SkFpdc@6Ic!)XYTM_3g(mG=UiT|2D<72N7ZQpA
z%GP;!toD{bneCWw8Zm-nLK2XMrPtnU_g)!EaXc-n%r{Z&_TMs;HCfPpB~zLq#2^|)
z^LA;AkK$^CnXtd&q>qU?cpjCu8mzgUYWu<e3eB6ff+YE3@`I99h-Xtj$CwKs-1V$1
zw}A2sisRYyY>kzxR;dH#(@+Z*B~8@gJIRfN0%4ujzMj(uaKaKk^WgZtkM?$nBTM_(
zDcE|I{z?2l0{P@sRcT}T*Xw?&YZ1qY!}~ovJiI){Fb3u+JU>$~6nez6L%nLuQaTK5
z1IWIk_r-#5c$2dR*>WOb_*O?urbNGGZ!a>$DQ7o?+X?2$*uSq_jHzb;v78QH_KO%|
z9PpCom%nXl!umdQ#;U#%N)0_Ijz=Io1nd;_|HR8Yo{fRnl$i_fZAMgaEouo)RZszQ
z_;-kym814FLU+6cW%EJU8OMUp)lw^Wze++Dl5P!Y7hqGow}{r2-9I7^EsUF!5fBi-
z!!dXS;4oAlmlDR#vVyc>Rpf>k;rd*DgrQpw3+Y5!2gW6`zw5yAHMc=MdO>8Ehe}`q
z3nir51&b*X!m0vWdfx83Zx4qLLKq#^ppGqsOtfwxfD4^Q!a|#feoM!PFo{o{f&*IM
z0Br^YOoRzjnw{8yJj~&~X7<0<NJtckW@9*=2B^rut^?Ymyr}jF<YdzIIQIe9VW3ip
z_wtU`?`%$=tub_R1O@UyzzSLW4u9w`M|sxUSvX`S1aRD?J8wD71cNen_=Bb|X0}$|
zZ<LChb!FQ)i7+QHihn9TMLPajiC0|Cq`4?)E`7)@L18dR?JL(s!J>pE=8rWnnofi0
zB22RD@icKiML-B?;ZB_V<F%_-uc9mmlk<k=qk1(iE)H7)g;IjXd+0Po5hkLFT9_k*
zzeug4&TV9fOcd*^Av+oJwSV|w6mor>0r*LCs-R#IK|#OFKz0ZXNg1;`*w`WXA`LEj
z^;I-V$Lz*DAvIL^ZaA7~Yq3nf|82+Cm2^B0e&jj!SadnIK#SqXfVXfYqz&3OFl8Kj
z-`#<vLb4vx?y(E5H0FF;?&cWAN<_A=F&w;=Rfcwj*qs>0y%oJBcqm<xU@BPmYNW*>
zwm?D`S^`In#9IxRqjdZV(S3w=sOO?VtV90BT#7yc5K6a#D06`WS_Isyb(WQ@abuJq
zv#MYr-NbAr9QXW+ii)>uQ*ovus4RI)1jB)3v2FHKhrEKpC7xTqqinM$1(krsX;>a(
ziKZyplnPmgEE0{xf>Y2W0gqWf2AeUArGEN!3Pp$^-jJ?KD<wWT+vAdV)Yo#CeHR)$
zV6eA~KI#tguU17AUqgb`E=D)@C^Pyarn)%jg~x0pR;}8L*j%*!eHK-af{`m%uO{_Y
zXUvh8H9-$A>>S1+lkD}@fWeh^hhP@soJoa*lzaD9`-Ptv>AE0i3N9yAetrS}xJJVW
zx;<X{8SajM*-EW{<z^zn$2;=>&NoL+w$&%D0!z!|{TQ%4_<wOhj$^pF><#B~DmUn(
zpmPLQjBz~R#P8$d+W*@Z&EBSuUmYRzd$m&q&^LX&-0)8*lJSoe;orcqf)r#c#8Y*W
zto8N1RV_fJfsU1RbaXt9WP6T319v_Ii4Z|Zl?WENktZBnvf@UbdPQiNZ?tb)qs_~l
zaem_NxEROkK=MKriEdMMU$Fa3Djiiu9}yl9NGuSxwbh?|e8;QrN2ed~`M<6)c^R^B
zWF|ml&PNm!2#$jCu%OxvZa_wbM1<2M_5ePTt#Rt@v-wowaPgumR@{q1f+(TRJF2}Q
z%MdkKYwK2bD5(&s`6(!F=QLGF&Q*^zTEq<i<tgk!F9S?W+5-f3;kYX2!rO6)q%NZU
zZ|+FmLBc5{=v%)YIfb+UCv$`K{&5<UY)<%&Vs&4&J_4<B!=HjF7zC{SPHyD0Qi25z
zKN?m-{+AA70|uaBZ%CsnSFct^4m|~z!E<zyT<!-85cfYCRzO}&jV`{QP#+T6(Jx;f
zgxj84T|cI+(%>C-IkFyRH0p!_;|R)0L0g5Y@mlJ1Q8i5g160znWfY@8!9vS$gj<6#
z-VmJFWAGv4M9=tR1*%xNQU!BbO-REaQ_zOOPnQZ&i4M3xDMO0!q3At*0M}(ox1*do
zQUK9*(ZOmMox>H-FQqyEgzC+3u5sJxASapwLQ%on*^O}~iu;q+MEYI@265={5^AhK
z?-cr>)Gr|G?AHc`Hs$3YEB{G$O?yB1`fV<qJCShmm$n-q(tE=BW7=G4iE(SuE7et0
z?x0f=4q4NIRzW9MQSu#A0YT_QIuFY+ouFCzByl=)1}_VkmSmx%b)<4Mos@7SEQ=~E
zCcH}5bij*4|EKv<n%u_4VGcpH-4SLQWjBk8Mr5W`-Q6{v;x{B}U^H1wNGMRuwWu0E
zk%TTfRt+922PJtU^X|ThmaoT`JvSdxQ4@kC@N4~gkZQOoOu(h|$&Gt(15<GpFP++F
z-Zl}@2LS=cJePgTZs5$dWq57Khc4?Ywm{`(N!*^7D1tMnL~Em8cXmoLHC-Ge`=ur2
z5p`LIs0cFNdFHjyw&78;)T5;){qn6SY2AS&h*iZ30SK<vQ{&q!D1CA#Yn9~w!A7AG
z)Ba2xtr9~g0SWZWLPSG1gsv|J1n*}7DO`kE4KU@=<9$-3_;kuYqT=?<6*I4mSLow>
z?ET5`7R^gTm{QysGiQ38`t|9}qdjjdJtxmF7t-aO3j7PLj+A?DX>MBaKjYoRg3RTv
zLf?M>lm8+fPTR|=7IMbUwyP*xr}YHGi!7QI#8F0mPR+ohsj<)dD!6Z?C3B%;k$Q@i
z(F#W{(U&vr@dVv*$<dwDjet8RJzWpFd!n=)(l1bR>9}K-Fc1-Kbc`F80ugbdC0DJY
zbK~6dT>iqERWr^)>K0*QZ^g7XR@$Z%YwH5pLI(%MikRD>z^|*q_&9!i(M$Ga2zD%r
zAEWu}6oPrK1Qf;whJE9G>c03xShFHb8iRw1Z{&jiBc-7&1F5?eOv({swThFc2y-Tp
zfEW-RHuE8^kH@o4(rbf;;amwH^m+;)b5dTvKKvvnM}Y-lLqq+lAD{Txte$cq;_$FY
znguGNkV!_Qkq)%O@LwN9PEr7A-URC~8U_bRsYI8tj-m!yb1(`5(GYQmYp|Pvuojz`
zn2<1x&<-6sx6W<Y8M~qEzz*C*%E>?oC(wqXM+r7c<G|Ozja?xlGvphup3G{Fil$?J
zptBJ)#fwoD#l(~W#uiHeL1_|c9Gc>Y5=a6ryy7!{6UU*1vIoc}iQ`|u^$(Qyx7TQ4
z-P3p@4(kE{T7h$_h+3m4so?_I?bmTtDCZQ(G<5!Oj?7pZMWAuT^f@@AxRI5x3^_*R
z8UP$$KrW_p#1K#0i#JaBF!1D<s($PY?FA^fngD~)KIkMKI$@h)QUKgiII@uRNZeTk
z<T1(HuIdRIZX7QUt02IxLPOeBQ4dy217N)w(Rm>9BKq$=EI4wca7dnc@C*6+2LQ3^
zsxfAOQ%GucDg=J|Nu|*y?Cbpc+uEir%{ykvs9Oo?9?_7(p&&KsN(g|#S2>Y4)HV2M
z>exv=diE6ANA$zAN5lf1gS<My({HX@l<G`K_8HgJUZ!#!4{1Z{K}DC0t0G|q0Zuvv
ziPcMI6iO(5-#STWY!vZjYFkZG3?08oL14`A#{1d;r2;MUAo^@#&y0{8-@nSssXEgE
zi!_P>`f#d1JP8$GFHS}P!{)SgyE}fHQ?|`1E>^-O;IyrAcf2Gx7J+B{zYC_k?2-lW
z5_ZFJrlV%Y-1l8jY*-<{@WjN-pJ%@^q87to==n?$Vz*6$VtQd22Y9^%Dh=iht%~dc
z0WgE1gTSz$;~#F?<3WR_;|DIC7g6+jha=WJOH6NqdBbhO#IJb*JRA)g`30*%V0wwX
z=h)9MHi%sKo^m5}W{q$QX>%$n*J+|3%C&a%;OFK*&uiCyVbYhe?-!|z#pxNfHy4)P
zeT8E-!T~NwdLkvPs3_6qC{_xM)lwLOj^-roNHV?a_oWJ7Uov@#;j=sl!RGC+<Su;G
zbT9b7BvnadRL`nAo^%nV3@Bg~lTPAXnm;Crxcjkod{++<pUEPljME3xD>ilfdXAgT
zoa%t#E@38e2bygjyBnVzmN-#8j*PnuXEN2(PPHsy5?e7(a0UEih5Yfy5At8<PjS0P
z?*v>4RtSn)8y1fg7n0v|XcQ}$L>&Lig;2Mr6b|+wMXO6oRsGr^8HIxPv^J%DCo-M|
zfQv&#3kQXD{Bd&hxdKFoy}9Pz@1<*9f~27cv<#ZZfcZ(WFziqfCfzX1caD-_s8(>p
zftI{6B>(q|cvNo=Lt-<X^NB{Cgw9EsuG2o%y2mm?>8uv$9+V3Q0meabEfK2UUcD^)
z`i5f|M5RL`V49g?kea4~aQPx=HfZKdswb#Z2npTupFj3<);cq1AUy$n2T@#AxWNlE
z=s+s^_`*GYYVOxRGFn$&n38^BmQXp(+#@Q%`9Kan<`n6S++fAA@50l3^qm2~J{&Xh
z)KL&_p<6_~bsK0|o`G>zz~1QBuV0@U{&ISJwYuxjqJpD?{|P$0{7pxHEvj1q5sR5z
z<$*Uhl;Aw}!p7IhK@^6j_O0tSUFe)uyJz9tu9=jwhBYRfhfR(2v3VvAb)vH_gvAXr
z^-%P|K(C07<<}ApepTLaIu%4ryj^xqAG#e95?&U#&yOsi@ZT&kjn8MXTQda*CvKRH
zUDV(@7!5SLYBV*bap1rTnkbZliw2Ob0(ZwhAb<|6!P!UEKQ(>54KLTAMeSjM^IGU3
z0Tqq`y@G*Mn7wq+4Om0D*`n;ew^%DU=jWy{xojgnawAKDj?yfS96sEOSdOm#n7flo
zvJ=MOL`6hGb8JV3`?Ext=<b<<MpfhI8DMh3R%gob3~Aw)*hkUAG~O1&8^BrldsuXi
z)c9s_FQ7yBXc~$JUD3VjY>8@N$Pe{{d&HK(-YEQ120=Vw(1<k6HW?Wi5t72LtQfn0
zS_kPQnfFO%8}YzMfI;)E_%z*`DJmgRes=25loZ3?o`A_x3+NX(G&Cut9_SJ3&Zs8{
zSqVua3Q=m9nMD@%Q37R<l>n>{gCzpQh7J(M;xM=5^ZoBqFXayV(5^_Kf~Z3aistH?
z&K*a~;c=^Q+|~J6LR{!V2q#29Y8iNWW)@1{A$y^hNegFpCBndGVV2mC)2HDk4bA7<
z<~=?M)owXF>q39&7jM5{d_jZ>A2jlV?0@8GS9y-e5Fn#NV;O*EMEL^j>m%v353s8;
zNJO0WL!BZPG?|UxmT>UFcXU7^a>b;X!bl1Oqe!fFp_x~>5!BbgQE>)>CX@LmonwnP
zXNGtQkNX*+ML9eXino&v29mmLpD*22I?5WeI6SV;e<Fi9opT~yUvp2wiGQpRoIVz<
z*@p1dKBo6D9S%<)Tp?gCI@N-#o2sxoY?(*ceb)VgK$2vlf)~eQ=oc>To|cvtQQ%f&
zu25c9?%KKYz_}osx6exOhQiRKTj{)GRIzZhLChA@nYl<ua5aYhFU$>{@=j234BSyj
z57q?!HjHTO?G^|~)FN>}c|y7c+Bo^K9VyL7jqt6+R5CL1Nd2h7F$$Ic5Cem0G#K9B
zUx67p<Sc^$wB8}fL_dY;7>xbua!kI*(vThfD?jL~h_*k69c;;k8%KYkiR|F}wan7B
z#akN3EB*0v?CNz<1{u}nUmy-{!Vm!kfzn-}no$G@IG2ZE@Pg#O#Z*XQNPm%s=ns0Y
zw=n(P5vlEoHLEJtL6M+?YJdeTGzK9R2s7yvUw~gDxDNXCHwXxu4S)IU^&vKP4UG<C
zVv&WG!OOr-Syen_p%Ly5oz;rREd%#Q&0OC`kr-h!1R}{xG^#~qM+i|1YBki>P#oP=
z(JK965alNc-Z&i^C&E0I(KkB51oe*=68T;zUX(4vS&!zhWWJt~4&m{yT0rSrxC!j?
zuva+G%==V};ZgYMn17L=lVPOga54(B6waWq{*THydybCf#RT<5T)N09ZXUS7O<WBf
zP-|ajtCjE>ldu$(qFQ_(C#{mAM+c+vbiU<trvJfW&Bg<5MOaS@<|X_8F!vr%Ri0_v
zHt7>%nIxl0Y$;-*Aetf;Y)OpKEwNJ+0eb_ngAo+WBx;H!?ud-Nfl9X_2r4E<jS?GG
zLBNJcw}A=*-*E!lB=b&Y<}3eNf7ZL!JCk9z=Xvh?zOJ)fMK4mxlL`>}9(5$-+)1vN
zz?jqNn?O+vuMB$bser^ILpuAy`w=T7c8Wc}Kd?kpT-kKLignbMH%;f~j74nZrPM10
zL=p_9TAFsK)WPry0sD&=hj5|1BE!SN_8r`IGAlSE<j6hZWQ>8_oFURNz7lu;5J7OD
zrD07A05eya$GsV1&ZeCnHjkoEdVh(@$baSQr_aAVBIUcJDuSldGq2%BPN6#({qXMZ
zSf38tL%rk4NvtT$dS&O;T?-wJ1g2^Vk1~Q-UPB;`CKQ)tOm{><rm#LXh^~9;*REdY
zMV!4i`4Q&;jaPmD{bp{ym_Yi3@Y^;N-Faibw;|1<X_`ajo{WnR6MJ=uJiYeRss0Az
zkE3I*r(4jWO1McJxj-dO5#`J%uDgEyFt4pKZY$R2PdRtQ>ybl}wf*d}^o>u(-#E9>
zG4{gr@%Xu#5^=y$$`ZVO_3B&HKz*mzYF!t)ZwB<>fk^~o(bqz<XKO?oKk}MqgQgt?
z{Y(0YsxR&x2frfm*5<ahPmkon%{*E0;Ne4q5kumLYYWg7TOImh$v;Vk(nH?8p9am_
z+5k_tA+v{XH%Cvm&w-{m|2xJ10G(mjx}mMxu{V2sUNr4Sq0ApI71x%o$*m~Fyl<?T
zv~NL$m%`VjBIKugJU`C;PuutQW#OfSReamnlCIFsqY@rwD`8|Vvqn)KAfR{Rh}|=9
z6DPlH&(Iz}%(=a9CKf(6&yYoNiN{jw#Y=O;QKeV|G&ggi9@4pL?u)}X=_KiWsGJI)
zJ!-$;3(y1KletefZ+VOSntTP;Rn^|?%g0}Soj^kQ$pKrQV}nk~k0qO|zfl8zZe9K;
z^Z+I}wv7-34GMb_iF)Yv@n(49q?)1lw<_>^nDdJG5%7YjbRSM`V_@Hw^FF~-7PS<D
zyA2qEk!`9?!eQyJKn=*Vo$9&ix8{G_giEa3J9iFXa2%0h4~xZrxEL0i?_x(HhmG%}
zm<gar1sozu^_hbVi9;jI;p{irotHdwkn>CfohZgPb5~N!Yn=iMn+@U;ilEojki=u>
z_I3I%L`1vt?JEM}AN8R<*(tD-({U$SbvSQ-9x$aH(#fZ}m7oEdQg~rE7g#O?Nr`yQ
z6&fW_K$2mUbR()+oCYHi9#ZxlLgX!(owfynOV9-kLgteKV$kj#uWMJY%HIR4Brg<I
zr|uqod<wn=YMXvWM4jy__Tf(CI-%S82R~&FciJ7~HMX)Wqf`?IKruK@qui!Ep)w54
zO}0@)(l%GiYHxPTnYt~?HugfZvNPRE!b+nQ3W%$O^T!Usu2f9Nj`dNVB@aO|L64!O
zU^e1zD4jJ;piQ@KBS9$=^}NPcl}5q}eH6~H2hR|+y4s_mscH%~Qrg*)?ZZ=g%LLBZ
zi-w6PU^`{akoKafVgokV*h3Mo0<$G1J$&HPJX`Ka)0M-8;bjY`<V`AErF6CH(El!?
zfOJc+9poo!3y@>)psQ8fbblf*Ii5+R8d)&u=aO~3UH6m$rm1=`&beJ3rhyXdoHxD>
zue}g68eoF>m~iqNmvCij%GbKt0KwuNi?Z%V#1#*PAPkBx4C;bpk6<ua%>uu(YNWdm
zZ&vNGj6392u$-WMQrz95rvDQ^<uK0PsxM1!+TKaO;UxFnW`l&mg-#34f3k39zateG
zFoqdru*gGhKKJLkM(II|hegj{Xm(84I!jLn4mh#HY!fEpQ}|xG)sZo@*<-k{#z3rg
zk8kji23zu5&*Fqu>Mg!RHd5J)x$=g@Y^-dt?OyEMmA1MR%_8g@x1sq=pSq1;@j1>1
z<*W)%R}7qWE}39F5So_ZY1hKH5k*cFgFq5#aRb?DphA9mab)eQ%{o)PzpWffnMRa@
znmH@MN}a(cpFCX>^!KB*77HZq1cJz;wp9vY^WD$X-HoHSuo=huR_4dnf7GyR;B(Va
zq4#`n`6){2VXXCynVnxX;Sv$+dk?&|5&DtZdI%?sw00;EZCT^{YIL@IthH;!x!#Ss
zZZ)s_ZZUbR2O$ZiSfVf9$z?B3X!cBpMM;P2{$OyohEvYh{b#co-RarjE7|uQ@6{Ew
zX2}grXdYZ&?3gFBlrzSr43$*(y5wdJ=bcQ?p@eSQWyWT%EX&C`t*)ON-u+STuGscZ
zm=(^Nd?d~zqm#-v3o^%My2!_x2fx3(U0si4HqD~p3!jf7Zuk(?cYx<<vI+Ax*9Kn!
z5!|@;{ld<;zF#Uv7Hf&T3GE43_$Fa3qDPDZRweST6UwB-`}vvrw;9LEruTA9Yn|-W
zZ%4O+*r(0X)&QMQN%c1>xD`#+ryLwqpn^_OSIO;}kZwYbuvPPun*R>!S>Ix25=@67
zk(+<LaQ@5}P-rPktw)bcw;}@aB&wNm>++8Qg{IgUvj-ixy08OAb9^<!mr=|Wz|YwD
zRUJ%uphzTKC&%)~o2a?X6G3l6J5C;*`2>-DO^ksY8KJi!r3{HQ2seKD4LG48>5S9B
zIOkD}=gW6!?F_GsiYRl(%V-Rw(do9ex#?kO%cBT3vctSB@tSUIEjNsT%nb>>&8F91
z0-8U7SufpsTidUk;(TK2EMv6V%nQ4=AgYLfL%X=Aic0Bw^B%X3JoUJ=HJjF-@6GNW
z4x-r4hy<@AMb~%hx^~m{MW2D<JMrZ~-M7#JyHWES1F2d}H>n13Iy7FHTCz7FDhWHT
zu3g(Z@)wEpDQU3fvIT<cwy-xrU6<xG?UH|--CBqFrz={9ODuOhGc!{Fwfq5KTS-ho
z31O#HhRkw{y7i~8`@-kpou-UFw2T0;){jj8-ubm9QIFF#x->BvI@>Yn@2xL?e1p6F
zT+7clJEVQK!6s$a0Gvl(S{}4vbe|YeOUuxLq4}RZwj-fOyxcS<q)7uTcq=bXJ?s$G
zPsub9d6l&{X0T7KJ;=DP7Fp>J?&Xz+&We1%kr5=Mr3Ia_1m?7g%9<<<65@1OS{i2V
z$lr8(S=HHkQM2uzm=2~N`Eb<N?k!%eEdnmz0Z5S%fP;~k-{?3dDGp*MU0oXUKA)%V
z`q2HXQ2u|HIdUCi2*R+EyrtrDc})8bBSMz}1tJ4-`N2o3a3YO0CPTa<%413ta;yg;
zo(L&jKx&KhROuV;<MIW|ZSGY21wM4S$w|)tilSoZxFZ!ces<-EqIs^4b<MeAfh@zB
z$e^%-_h&3~wr`Ym=gzEz-DTOnWyP89T@Uv{m$87Jgw4Vp;f}(G;d0#BXjl7S+E&(~
z<Nc0C*o2N~_9V5?;{|}gIxJ;cNIskGfv}f<0(f(nFd-=OEE{@rSJQX91sTBh5=>zo
z920Y)I9Bk|qD9JDP1|i_W*=d3+2h>iFHR5QrAmRJMHj1z)WV)M%+vynnGjOK8))iV
zt+WsDazf(=u1dDZLquKfS6xz_N7tFiN-t$xLtDCdMRj?3d4-pFQezrwhN0kCciXh0
zSJj0pbqNc}{s8g)g@SpUc^z_4tS#-bN7dMQwEsM0V&SK{`eA>0t^Tv9__N()_T&MC
z=KDqpa#+s=m8EH31(*H|*co!|E_yOsj!ba202$T<Z*j}L(#T@6V2IP)OEZ_q*u)*j
zf^w5D;1wAR6T`u+eZYJQ%Zl#;MjPxE$^N9No`8F>2E&lwYMbD3b#q6kR%5kl(nRIA
zmMAM*MG%6V^c%LR5&(Sp<LKx<pM3V&Qh>~qEhp%^+|jDg`JaQ}vqqhn>6PIj0!MoX
z+m%!PYv|}H6BWhz&c@F_hKJi#acD~!_bqA?W+lB8nNl*CNz*h;d3lk3m7(TSs0kcP
zpFLXrD7Tja-UWW3tQmQn*fdw~yw-+9<^&BdczK`p=}nt9(J<v)zgK-R?(VLTf`ir9
zB4-aC`~2V9Ie*;x#Xc`aH|c$>*_x9RK0f;XhK|W?X12ZJW%%L7qzMO>-RL={eb*7I
zPlX3xOW*E)d1z|*m^Lq3xcwz!=j6Yyzguy%VEss&=bpL!^@)uu3a-XZJiIBVqN-(J
zFW32>`!(>2Z|+j}kWQGAb?^Y!N_=wMHA+vTIpo;s;?KG6;;%Wr(ocU3HN73#!AnI?
zN~1|nuTm2b%#EC9D)t?$V~XovC$5k&2wjoHOr*|7Xw3%&>5$T%49H%TO7KT_m4|W^
zQbD%@8(`5`D8+5mnnq+B(gryx-V0_NC9Tq5s8hgD!Q!Jw3`?E#4ax{Wrq%{{fVC9H
z$l~nAMk*&Bk~dW-{MM~+0d?IZa5V{Q=XJDOk}mvX6MzO7gK`T9)sik#9WA@$&{2CF
zoSo-q21F7@k$!N6B}F(T3bb5Jyey_4_sNGL0fC0y088CivursMje#M!oiw)Af?($F
z)cdY&x^V#4m}ya~MSgY{T6k<p=Y|s|JS_T2DaapqQE`QEg@ucXr8|0v7lx_cBM~yO
zt`6DxS#8pA`hnRyGCMU`T~?n_e0$=u=p87H8F;oz(S2iorD<aWT8V%hXWAQiZePEK
zvIQBAR9xa6xozSCxGtKGCR#4crw|z$U7w<7Aj4=o6nP>W7+T-KA~Q9_sT1v_=uDfL
zSHDu?bqmBb2<!aKd5|qTbU16%8PNnT9ycByX$;ZEsVG9)z(pgBJCRT=m$B7%y4$iy
zYlhEB_y-ySV>~9)68{rbLjX*fFT6t~c)B1h7P5>Lz_3t*zrH=+KM+w#?R270WZTLn
z%RIYnD_D!<FH-064a5ZDkWvI&DHA=x=JqWY@<~;l0%XIvaP@ods?R8{i(zaMQv$yC
z@599mtB!+e>%1Vl1|w3eX52Mc8{?ePv(HWY`A)_Lo+`lf-bDMZxJN5@GrPJfnlHO!
z=%CA4-!?^=7XmYSVLlJcsm#pE#Zk4(>5|3jPAr6}ZDv-2K<T?d1xCjCV%#!DR6VSY
z@~wI}Nf#4}zyV`xYlEIljU9{lB;xv&D`PzTr*s|O#vnk#Tv|d>Zc__*7<e2}cW$HB
zS+#QIm|?g-*KhQznpZGta<lr+N}N=ux~AX?n~gG7X^Y|p2|jLJ_RaF;`+ZxcS(NN`
zK8q{`l4K`9sye&W0vrJRkYo~rguddm%LqGt`gDp0R8t5R<QD#giG6{kd%utHZNwFu
z&?2tv-9WVCGI|;;UJSf;!*1r!+xaO;)a<1g6;?mEVzDeJ`Wtq(rN`HuU2oN^pR^{K
zw}BS=4_!0xtW(uYijD$0&vx$t_*-a17rX8Lhg%LewwCd*Q}y}1w?KsHfoB~puk8`0
zEn?GQ2%w_$yxCF1(ip8X%n%ekX_vC*L2Wdn11<2GraKvVFx;r*X9F}rfV7stoC;d)
zRDlk2uxx|=7LM1%ZsUbWKq4kvQKrud##hSd$Ev}ry6vTLW$8WoIX;z#5OoTLK&sgR
z@NM<#r=0;kVep0*>~2%C_tGyCh_Y4UCcW6A1PlU3VF%B5#`@tcyUe)k(e>TQCD~s@
zN6!r&$L9r@r~f1C$%b8jbf4#bRm)`ibQSpKd5bH(r(dU4G1}!fTGD!H)UR<Hp&$%K
zU}kHN^o~kON_{01@}>ZCQ#jt~yP2oz1RWOZ7!L`E$VzWdOMA+D0oO^M3r^q2ntek+
z57bfVIVXrE;E0;hg_ID+s*+tUQ%IZSP#$sg6taK(uWc#PihYcMp&x~qvt1sU_^azJ
zi^}X#fz`WrOyl*y3DzKD<nQg|*{eH?Pq)lju)3ga`}UGV09=6xyiNVGE)ti&p{#oB
zL0XK?2Tz1w&d>KAHg0DZ8f6yhPEFnrRY}6_c`BHqjlWu>l4jgO_=yvokojNF>VGLR
z43tOCMQKyu;8P25{y;LV8o050Ff-Wj(?&b@YBM4DO`4SW>Z-WYrD9^Kv#jpc1;$NV
z1YVmj=3KIPic7@`Na@CsT9bO>=bsge$L3WtFOhqD6V5~=g!OtWWc7Ac@80tEYcTD#
zxLG;ho*q7B%dQ_se-GEsLMU~jPv#I2T+WRzkWqz<E1sZMJlWahuVggoT+yy<_#Q=_
zs^?7G(2S8Avx^+|Pq@6VXv4`jpO_AkMhH^}@zIpLDP+@>5{Ra{QHi_ALKNZ(cyK;_
zbOj!n*y8Edr+&hLu1o`xehaIvMIAg6zOU^0G>ui)t-a9V0}41IYh;uZXxaM7tdT={
z1Gvi1yRHEE#9`XB;s)-uZE=i0tMhFv3Noxh+9dH`%_VE<kW|icK%!}@suana!W|8c
zBVQPP3Eqf@Cv#xgBQ#VYSYh=R{6>RZC=)uZD|&YS>iwI`{(crSyb$_MY48u0EIJ}N
zhet9)!BM?6*gNN$-Ppsee?8<dZ&QwUh+VLzp1Cm609ioLvq{lUdGs#%3<%*3gT&bo
zJ+1Qc8zsphLteCU-zdn3fNz9dnZ_q*qE@|Zm#Ca_lNKQ9&e#uZzb@~~<jqy)5~Mdi
z9FWubgS4tCKNUsPZZDU;-UV4;n1;A{NORE}VmlSOdZV-sBt87_Ml)KO&p2AFX5LAG
z&SAe%PL#);CI+$1luT+0obDImkfGqY#~yd#Pzh2<a|4gmzSmHi+%Jq6ftJu*xR)_v
zqf_P8?>bQFPkFhxzvIY0kA`pwK9$InX#zvo;(o}yjd15f;Qe$)rLHNnT-LgKd$_9C
zx-kJIDyVI7J9^7dyWp+z;8V7jXJl7RoI4Q0cojxQr1T?aIoCAdc5cnS1dOZ+TZ)KV
zL8cwme?-NK)H{tTSM<F#-M%jJV(v?}m)8LQQ?e_L$hn}RwBVgQx0X0wWMEp?Rnjl>
z9t^5ih)Ah7T2fN)&`uEiSz3;~r>y~IwEfC!jQo>tmPb30=ESXNh)+26pCwUUKS$M#
zv|-~&0>?0SY0#q}B~&zg3j{AeKO&8D<R~{%L+rLp+Py1nf6|>1r2X`PI&6KM<*Kud
z^tpQXGfHJoVO{M0`Q&w;jbH{7v`%-peCEoog=t&i+U3Ziom6q4i8oY5v8-QkQ{_*M
z^y-js{^8kP)t4p~LJ2Cy7_q3w$0K}Xpa*0dI_N5ghQ0Bg;I6P*Bx+%|J~DzF^7fQA
z*%<L`K#ofZAUH(}84n7V?KE2>6(oBDVqlvRBfHVJTtg5hb4b(33}Jf+yUR%v%n7U<
z1tRP;&r$^?2dORXsxVt*Jd=}Bk06xh?hD$cBGxWJ2-%19aWQl;axiQ?v^^;36RdGC
z3LU))hbto$>~JvtoSK{omb|y7sM3|J^KOE)n<c9J0md$!r&S6chel07Sv1;``-<no
zSId<4mEc|-l<LJbun(~dYvxWZry1iO%r4+Xk|HY-1h>Q}7$T~>j~A6K8P}!z<*1!Q
zgEZhO816iu(MhxWU9E=b*T@Eq*bFOrJuoyrY6Lw32r5sHhwn~t9F=%dQnSAw?p?^x
zi~(!j@55l1_Pf&*_R6^y!k*nJ{)yO&bw^(Dc*-KmsMT|#c6dxFX&;+a4zboR)_ZLl
z1x)clrc5Z(0=#<|QP&b4%8JO|Ce6Xr8_H0%vKoM^<=hufJ+}{EB4_g;ZE{q_^n4zv
zG4%iJCsO`GTt#`8x+!;Cnm{n&{`a!|1Jx5kfD2ErHGp*2P3ASCr5V819$62RkAIi&
zky(>2GGJORKm?KnS3dqu!zpPq*y3y!l<(6YOQ%S4%b{h@hfIIaBW5M<>IMp}`=<hf
z^;1Cac#e^V(j+47bfYd=fBn$WF4tSTr=jxLU$OnCL=1;LIxj#zxS38Yk6{?8JYy89
z+W|4%fpD}{;k?P*XsIX7yz=g*YCE1E^R~d9<lLmuH-!pk<Sa!YPurD76R6?5IHS^^
zj_N#;>F#Gzf{$if#yt(Xbb0?28Y0E??WUf9xiom4s0}-qoq5R0%7!P~J98f3JZg#w
zuZ0|v4V~Qg_DRkO&yUY@%Xr?l;G1U8zwkm5r)SY}xEnMo(aD|@7OZA#MV!4Y`8LlH
zkJE_Mte?_eM*f(8|2+$fkmTKLr_sqJvx!4EJ>~=}dgy(Gv>iMPy-T;uM|%Wo(x0sx
zJ+u9RRQr3)3TWfiYIf~9LR0RcD}$+MWzMHcvb=k81Xs+_bzUEh8`1oyO4R+4nqnk>
z5{5js@OGuqjiW=JWvKj^qR6;u1(h48jkv!*OL0&!El85wit`_u>csY()(h9=q9CFI
zl_HlpF~wgOO_xPi^Z^RXgpPiEYS8kF6{o*yRoC{#tkdxVeujN#)%njeVi!L-w+H<s
zMOpOXI35-6D~n#a4)Xl<$hu>zVJBRBN~RN2Mmt4nK+dW5&&PxclCQC2)s<B(X4cIH
z>+*%xtMDr7;8HNPQ)r3y*gjco(XeZyO&TTZ%~Y@EyL;5)hU<T(GhoPSW2x6!<vR;!
zEsR^<`$^Y?>0`xjnEuY}!-n-fudYe+qmBFHzaFJlArWnJi$DA|Gw4fU@Dzn5aEpqe
z#mr{4pKA7CSm@QS2Xx@tvjb~byhgT3YNLs>ltn2Dfys<gv_Xp@FV(&1k|$lBFWa&%
zQ&@E`UPkHBKFMidMIFEK;l(b2eQG;3Y(Tf0&-#03Y<#rT%P-XKDwp@W-8=V`zsr+m
ztG<2Nq&)q@+{Lb@H#m|xwz4qT-{s>qJzlKtKOw;Ea;Mq-j*KK`l+xwEr2_Wm5L)qW
zo8PMYdNCE^4e8xEkxm^drpRl!(?o9+4+PPcTOHp`{sT%nO~JBaR2{y_bAqEJk)~&x
z^GVmbcbnljtKX3f5YlF61c-ohXcAk9Uv(8_*y<`Kf7|>9%3Ho>dG`jL9=o4x3(N;o
zEQ9?-)0<tkv#`D#^(xi4`4j+if9G1)5h+M3y1jMz5ZH>jLD^Q<``3N?k$IP1e4>QK
zWUdsJ(0}^(IMCnFwE3B>9Zkg2nQh;;>eK^S+GlSkHON(QCrMrHbVWI&yZ-LW!XPwY
zf*+Fc1F3Z8E-Ed4Q((p=wm83)+NGgr8GtK`xWAb$NmDN3+4WtwiS*1X;7l(ESzsD;
zi@vLBzsJH^U6kMh;0E_rT^?CY<|^mJ@U$?OxYQpTXZbQ+Eybl`>I`4X&C7t=Uwb^!
zEWX#d&3kGEl$RsO1_%o$ivC73h8BMch#qvNGZTlo<emdYWF=gm+HTXy%m=(dbX1?-
z)J7V=2PSdpOh)ZD*t{N+$J{{p&%xq@@Q+yAhvz;Oa|r1`6(qCPo<<RiYNB;lRW06K
zK~H4PoNrzGOE<44XRtlzIC_)&v3IL-{i^yJ?-lMno3(t8*`qIxKA9!%AEiD!E%z>h
z6*g?_Bih6A6TPaoupdrAM6leVZXtC35ndfGIGOF1(HD6`b0uS;=J2K$pN{OX9ee3<
z&Kj)S#)z`(EoeP;oZ?c<*MMD8Ot(;;*cwF^BI4~9Qg=fxeZ|-;TJ$Np=t4{wT>+B3
z?zXTxs3VPjFO29#A<G>>-ZjWSW+PS|^=6U~fzoJZiH24t+kEi;`+s}om3RF7{D9m<
znQB%w{?W-F%w-!@9F)04cM{O@I)&Y9ObQbZ+m`k$n@Xgr2u(U;cF&s8kpzAP70)Kc
zQ6xDoQjK=ult*tC3gx!$fWUAJI;+4$KzA*O8fNvaLZcwnaSfFPACC&574F!F<>gv;
zX_cW5@1!p6whq!yBm%ggF;LQY_1&^p;q1zXpiM!P93k-`Xqki+=HjOvsX9#0_>lpV
z|05<#qE*Jy4xARKGxX4erTt$5v$nrhqvR!dULAcf?OHjHqmpw&Vq%7av&kR=(wlJa
zVf}xP(RO&Rnl@g4PSrsW;tm{8U<*@jUp=Odiu%3(uV>Gk(SR<7%B4UwDOFp;^r*h0
z{tJ}V`O4avxooxF0Fn(uIsAK;-)zOTytqM(Iv_(H8{z7}q1EGR56OiYXD4^0HX5kB
zwv6pRCT`q9k?WOKH?|Q)bt2X;+od3lhloBDnnB58?$YCSv;RkEY7-t~h|n6ghO4TG
zT3Kdfkhk$nq_~QxxHz6<ZrDozI8|F}-eh;ib%!EF0omBbZNpw8O-O2YIO8@@F=Ba8
zIIAzMarIBW?+QC3`Mzo!NKe|S!B{M8Q259U&AH-q0`Z<w{1gM(>5JQ^lb3ee!~@Hj
zYA|A&ENXW6qw|wHDvI0~$Rppyv@c&Ww_S()r_-t!-|N4;s5hn_NpqS0u%L+6b5%?g
zB16q8lVu^qA>6{ylxzs|DU{#D6GfdVWqKNGVY*>YRZ5IGee*!$J}PSDi%yZvmVGJ(
z6kr!Zlie9ro9RC!{f~<J3act%ZXJQY;NA2Hr7eh{O{DV8SMc$U9h1J<E&0WsDf32m
zeeKb&ADmqk13nZ1?nlbN`C>z^l1^iH<UFjwE`2Bt!p9+j6t!h;;IV8R@!H&;lvt~)
zUT4Rfur^!q(vB4owHxi_GcO{U1{PVz3>JAM{IfRxxRnK>1dS<4qn{tl`P{~!C~ux?
zB`cCgn>KM17m6Ic**X@?(UjH%1#<z$5ubKHtJSAVzc;6cIyNzE-{|z0`uXvt_*p!Y
z$^=xle1B)ol=R)>mP8$Qhc7f3c|Q$le`%a{F$k<o*W+b9$4#GJGW0~dl0)U=hR^@N
z`q~-l&9s*;xJrC`m(t1Hpt--lZ7R}EaC1N@g8-REL=q_47sV)*K$&OFVql2?ktB26
zzGX=Djn)Wy>%M!|ce1hy^A~spA!=)cGU)(@0~#B*W#=60zpW1B=(w;HN7n_Hab3GU
z0`hRaf1usjl{~@FYL-;jpxdjVC{R2UovBWXuL^=Kl873<bES4vtn@t47uN4Tv^q4j
zV8F7>Z)+EL28tO&_5F40n^DbFwaPmlf4g=?A1Ru2(i<4BgF#v?4853T25&E2-Ki0a
z#TZzJ-U%W#p?60TC%5a7wPz{A6N7%O;1D9d&%H;GCW>jeJ$?>xyKp;6!1m02?Ia<n
z%QdodsK{$(1OxkmQw3^sWUp6(Ga3Mw8#dFAo0gC)6+C5%vfl>~QE=w;FTb1vWW*2f
zY=c>*!foe54LJilQViWazmg%Osral(jb`K^;f<-C8tA|_4@NSa+Erp%^C|R`LB)4D
zG$5%5B&O`LGm5*8o*r(FAh0G`X6kVcqoN2b7L0U=9aC@~s1t=i%|t)t17W!q+$I8r
zzJ6wA6hCwrGJVYmS3n3%EWMs|xBL6g`087>ga9ghjIA1^tDj2A5JZL!{?HTUM<;e#
zr_54i{&7r5%*|Ut?Sj<SuA0G5(3uNfpvUzjHb>JNSn&=5rVsYyDI*j&jl}eAXo7<|
z%Jm7J%NgAyPFcLr<Y&gL<lyDFDM)b~6;*r@=!^WGRp|qNOpJ_XpXfA>6Ui&B9VDdj
z-hF0yC+EJm?sR!Aloremd2uPzGqZ}L1T_xoJ@VE1&$81fM{||Ad$aFG7EN?pJbUNh
znrfZXJH68F{MJQONg^W==7sPMAwU?C8PPn}U^1UB-PLj~CY@D;;mxvbS+Dk|Mmvyn
zVs2p9)!3Qf4s9(507RM_dU)Ua{^a;Q8Oul4=ubp?6T-a96%(l)SIb;o`F_{vueOp!
zq-VG^7C;W!yJGG%PSR`8wf4hz9cg;QX+FKbv?!E~5fuR{;&t;dxnsi~KDLn8&`uc5
z*&atkX44SU2quzy7lz*PC3l+F@jY6w2Nw)28Tn%8VIQpTz39m-Zgd{Ie~@n_WmfAO
z>pm0xmpZM{5A63;9`hQvt<y?1OeC;{XgL=-cmhM;oH{Sf$H%9D*Rq!=DD?Dtxz&}l
zZ!oMx)Fk8!QJ~`Ep>}@&1(5o>;1LkA`gcujrpQX!()&4iF(}j$S#HYK;PFUb@O%dD
z&r+I!U`oIm!0V}}N~IR%pya5FACvH;?YeJZrQjVVqYf3a#`C;|9W4)(DT(iL4|qw_
zs8I}vxihaxBhd;+IjNc2oJ?!gm_2y#pxOS#JHNh@T0eJi&|BZtwqA2W-Rd@&7N<A^
zVXegg0NMyqU*%JYmsny(4(34sBVM_}sJMa3s5|`N>I*<D%0$+|r!4OedLeJ{dzdgv
z$Crj@$SZ;UOY9Kx%OP~ID>5q(_t?vKrtn>fKiC9nVj98!+rII}<;7?3{`~SF0a?fu
z(!hh5Mi-)s&aNQxRq2N)r*NVeycGupFsa0E#YRDTZKs^jSR%;YByjU>0j}N#q1S1X
zoLqStisv`3_jyriLG81=@GivmN5LJe)=^C(%pE_s9h)~PL}{6Ed~ahQO_%FjSCAuO
zLPsoii<AxqmL#Ej_EyRV7~E?Oa^)V1s#x>YqDB6(J#pAlmvu_x4W%#h4?ZlACjJ!|
z{-EVB^s8*VbZ*`F-jqpVmL_~wSvD{LJ-}W4T^i0vi)mHU&@i%DWwyUd-P&tD8Vg08
z1Y>;1tQS}QZHD8{!QU<Z_FWE7dP7AVsOsV?8Uy8i-Bag<JYw}{sL7?Wh4&%Xlz?SD
z`^IPJ!7__8YTKmConqm`JyHMZ`f8J=T(yka2o|V`_40cs+bkTJv;XOX$VASu3dDKj
z!K3bejc~?T`JQ;JGndzGk~7vVWkxtRHL~#iu$I0L#`Jh}>!?4`83|q5T|NY~t`-Xd
z+YP|3(=|$#{OtI6g(xn;e41rhd&|rQ4~*`3*$Rh`KvDt<Z<0%K)-!h_YARr&=sR`F
zWV2g4hF)2ihPwsuCyiZVhe!oF#MZ}`eY#CyI?|V;r_l3h{$ic*-t5C??Au&kP!yFW
zaR>^n!#{gs9n=En%&$u%HkV$w@qrd~U+8jE9w|!d5HuJD$KILvGM##7($40t{@Y*b
z{$^Ln>e9;+4smR6<`XkRtZJ8?nOWMHuwTZa)2Yl>+b0DH>H-D$lvD0mI-696o98sP
z`O)XjbpxfvKrv?E;cX1%binrzK#`yyq7sr~M*@>qUI`dE^j)*fv%bpA4zF$GEsni>
zJJQaJK?Nc8Hb}0UNF!U5bFIFQJylaL`WZo|dp^3=Tk?sltSs@3>>b{gZYI=b=ii?w
ziGm{9Muc<$J6Sp2A?b){QTrR!`6~^Nl5O{yY<!+jqSI4~>VE0vLAk3BjX(Vpv4w!9
z!i#03rlr;t9+`^xY)@AJcMy(y+0~WV-uQk>L9Vm%ht^(3i+1A*CJBDB{`RKb#nxsq
zr>9+f-qnY5Ldj$rF0Zu#<lFeg7ss^dg#<LqyTD510fnyeiCT<!EN!%J-t?Ah0!#x(
zMUq4#DA81*u`TakPHNLxvRt!^Og2rk^;sS-d>$!P6yv8q5@gKQpqYpaPu})^f78v}
zVc%BQCoM8k&K^RF2~Zg~A;5&c4cwOal&dtYe4vAu$?kf&WAQ&8Pnj6p)0D^Z-wdxX
zrPP@HVNivg|EczX)XfjB{oH{&&Og(>1ov)3NQ+U95^Xa<XHJtN$|DvD22w!e!yeaB
zI5jovm7R^GfW#}-j*EL8{otprGfc5RYTszS6=AI?7eF58xxV<4P-Pbtk9mJhf9JCt
z#W34m3GNZ~gL<@@5ofa5ABjNq;m(7T<C;dkh~O3q@l<!-5xoaq%oE}5b|+r&`E>L9
zrtF9#$0@AktCZVAutr+CavlYXa-z1dbIGarpnI$JPxVj;cHx7{FP;^MY4%}|M0D~X
zDhpTW2&vQr+D{00=5nLQkwS5rFByUlLmZgy#tlz2EE@DwddwXFSpp;TDG*7T1YJX3
z5&WfK95%>2U^db!HF*<;Y1r6o;1f2MlZ3SbV98Dl#@l0E%Og!Rg#tP?g<ni3k-#4S
zypr+O<~_|j`4b2yx%l}Ty?TzDw{ZgGrhEciwN`lD6%@op>~P;O<-^BuH=gZKexT8?
zr^MpOOW1c~MTa-O`=4^L@f-f76l~O6F)naYszN1W!*v3k;E2Rl+b$dcRnq=X8qgxy
z;p>zF7^Qn3tR|4-nq!M*$KPliyP*O*zwQ!jumXfi*%S(Zee7#>7fNtv*K)o8T0(g{
zP)vK)xax3>M!c5=$|%`T`2cj83HM%ms)=a@wDG6F0#M$;%s>}tjm}Yh|4Cbw84ylm
z3^Ig5+=aeTnRGSTgu_q^L71+>w2Ks5NWO+BB05=xZ)E4roiF*w<18Qqzl9(#K@Yr#
zBzop~$6gt=?|>hn68dn}j&%BI;CxYDMNZ67rHc~=Y<c&C44>89d<d^#j$I0q^~gk`
z;8EzM#JTWvYc;GXd7_``H^qE`tw)0jHDycL?n^Q?WJ|LV2o$uZ_AV%e!!j5hB3;9A
z(H7|ZVKb3u3u|<1rw+phUg}BoT<hW&A9wv^3m*7+I8%w(>)vpW9P{P3y;l~Sv;aSD
zhESNov7mKJWng))9efM^qngk+^1ENU%Mh;C-*jrwrt{;d(*VRZ!1Xn$cotE)9({CL
za=x~%nInrz3&_E{aQBJAh`s9@W-$@6)QKv|d<vrq;M&^07Z2AD@4I{66lY@TGAo~}
z|0-fV%O&4-w?%NNWC(3Sts{_Q3kx$&b?Rx>`SDP#X3`rfC-o-qiZ{n*i&-m>vtDrS
z#-7eGfi%YB$r0ay5%f2bZ87>#88TE`U>^ZNS}xR5%H>q{98#6c1@Op)t_Z39%|#J4
z_HH77xC11q)U-ghvd1YJ3oh+$){&^kAs%n@gb!n>TF<*PUR-VT{A%}%_aCi1Kiqri
zgc@Z`=~`62Ox@h7{S7dKPya#kP@2$C+FoS2bJi44o6|n<k?rp`^*L(<A*5>#hL4c;
zp0a-jC6&^*hzPU#<L7H>;~X*%=J8cQ(86~RutbgC#KZfB)A+~ju-*mjpdtpyf|1Gx
z%L}ZBo5y#fS8mwgF}IqwmuFYQN3IW%bCXtj3JEaM>nf^5S|QB<Ca&6IU+1cWkFUW*
zet(jR1@_W>W(h1gS~96)(-c#K`qmiQ9AhBHgZ{DUsarTaaGFj3W*%)DJG)d(txw+g
z+pM8OjhU5fq!xy9CLjAIogLF@T`BAA7lMrFlJ~sEas6mO$+BLLpTSiLs5lS^Zh>=I
z?NfAmeR<T=d;X+(OnrI36nA7?-Xm!M8A8y{n!1-rU({L^Q~US!V{v!?hx={1Ydxg1
zX3jjn`J*pSxx_63aD#5P8$5WksP$L7!A&CwHP=bkc|oH`T}?;WW@pOZQu@aL_hsmy
zxb1)vp+F#$@j_2<&A?bEPXuJPKT-w_lmO=7VRvfUbylG0bZiI(winPa2)#87rOKsv
zXr<>=+6Ee(?R--N;#u2se)_j<VdRWvi;yBY8Z|SFr&n8<$SrQ2a2=&-Q3M5O29q}M
zkMIAq>GABWQ{mt&de#o4DyEiy^bAT>5fKHviK|`u6mfCk&Ig7H9s<>4CM81>vR}Fj
zo*afpWdz3<^USwu7iW7_5~gs>{hl$agE~y-1W?>3UeW??a5CnQ6!6_yDeh-Q>!F)v
z*QW-yC+uJ&c*DL)XIrYR#Bv^hf)b$~qox+s%WmY?(B2}hu7JMOm7;1?e!{O26W)Dw
zYWnNLKk3uwV&*n_Pma@F2U<drZ}(aJ)mM`NOgpcibRJ9-ieu-mcU!BRscqE!%a(Tb
zUR^_)uV7dO>;2+)&-`qP3o@JGWZI+uSb3diCd%Nq-Tg8>N0!{5lMowST9eFHbx`CF
zw-`0$wmr5_oLXk8;r!e*GJpo>%Bl8G#kAe`*zod)ak8erHe3H&{&|<*vLI&+{ny{d
zKDe-sU@RWL>@6F6RrD5*v&#Q;^q47AC~ao>50xsRANFkKTvI|nE{Xn7LjG@kC({Zs
z`_qsA_dmAFue3bkCxO-NyxJadc|4i5@oFT7#GRbJ5z9a!N=-Tn%Pkg{i~E}LUh84x
z!3J)J{0mz2@}MRE{-ZAZ^2;xSRU1@R=H%rmsyiRM&M>4J-^}|LWnW?OJZ4YIog9K*
z8J*r%*0Y=^$^q_HP1{82w4Q^T>6kTRkifpQFiu}0UV>Q@!rNo#Kiq?WTfSk92EFxz
zS@BfUy=^lFhEW`mf{@t4EitkXn4mrm3`Xg?1((R*kcm>WAa_813$f_rg8)h;hnkv{
zy^{ucmSnk5%-RE#S7ldM_;tD)Us2qR_*A?BrK@Exdhx|AT*m=I;@O?BOFNJCWs(TG
zFO6^2ilmm6Yi__B4K8_@HviFbda1U>TZ|S>y8{@DzL#`6#7Xhid(HrxYM_+4L6(Jn
z{dQICi9$D%PhTo!k;dt8j7ieK|5FAaD!1_)){&q{>yg0zFVUZGWvQcvDRRLiNd*S@
zTazXZW0&=zutjJ?X($ahCh(u1*!0qjT+=k$g|a)F8^~>OAT_!)_WWc!ST^v*oysF0
zKYo>awweZoyR-3@Iw1#uujstyv+XZ=yh28GI5q@fQ0&}$T|2*8Uk>!+D_)Nz0ftEI
zi#K|L=9F{rH)f}0hh81}>5<n=kNby46oR63@GITieWK;wZYO3r%_UtSC%uA=uicBq
zPH_2|H`0G6md;cHagBj6V`*JUl@ron#>C(4bg%u6Zx|D1ZjFB1$bS6AtfCfw{p;ZE
z^SB)Zw68WVNXV%AW&HaFBlOnz4^n}fW~DsSrJJdQTQ$1l>Z-okt9L8A!~JaHlh=A-
zKsEl8BzM{V`VaP+W=t*&4BZl9E~Oks=4ox#k%WBGH^$r2E_i4k@J&6)X!{eI*MG{-
z!4zA~Yj(^=S}Tr9wUt-yuN;C##Z)gzxec-@FbW;_ro!dQr51n&M&82VsmtviKV%BR
zCpaEE7{;ptPn71+DDFII-XymZP&syQ8N&Yf*49f~_He36w*bFi{i?0OFr>Ea$%}sB
zYty<<WA0iIl~)i#7|qW>a6K478IPV7Q?Z7k;^Q<J_f6{aAod`S8tbKDFDQ;3cRmZ*
zuBQQf2Q90@{ZuHa@v8G~2uQWCzuB{%sY%`Z6)8LN`(SslVa}N*KhZ^Lt`jwwGE+{k
zg+gr8rcFO2O=(nymd?GEdqNw+D>o1tu5%AjwQ4ZQhRMQieE7K;oiC@=K|i}*#3+W-
zAYh&^B?G}uV`ldtM~;#|*^}vciQM?Y<r)f!N^?(H<)MvVfBkhq)^jl<A17n}ZyqtG
z=llP3J2(5o*%e~vRfmqueLs8BmlDvmY5dsxe8li4pVp{0#cyFjoAT^dI<V|LMnnT;
z%~<#UgP+yrP3|Xh4zGAsZLt0FhOsXFUEJ1)=H`E@QGKQnX88_=0GJ%Ru@gQ#_S;j#
zwgZ0=%ySw->d`l{`AK7{pl>?0ZHnLVIxqq04SS_Ivjq+z4GyF%?H?902Mu{S!4(#9
z^}9S_TJ0A~X$}g2UfbUGMB*WDy0$B6yHV2Ed99hib7F#B{9R5ppcQ~03Q@J{Qg6MP
z$%%U^2!XCsO=OpBRu`v6T>o}j;g8R~o{S#N+|RH2RQqRQf=nxS@HDu>Rpt_I19j}Y
z?zXVk<7pYIGNN{PMKUO6H{7C%TydXBP85_?2Of5s?HD#(3Hvp1`-k9^70&`;d2C`k
zh(Gq~Bmq>^ic0;JfbsM}7Sn};<BOa)?NWvf>d_jeTF0EwS%OG{&yL!dYdI+VUmWHA
zf9*<TFD3#y1vDlUcz*4B;}!#2PM&@C&yMR=PVcOL;qk@s04p_`^Bu9}1iab|@fZv*
zgOO(`-4p1a&Pz!xbel{wa~<mo{&D(+a}g8f<z>7sN)Q*Pw6z@ViuD(gOV0|`HT;_c
zWv+WBzV~N7!NEY>ep7D*+<$m#m^;z%0H<vXVmJVo)bbZRpyb!oZHU+`l}Xsc?qRf<
zf~yy$bsE>(_EKjKU_4dYrI@ayQC;BMu7D<9z1+LJZhDYbjo?ey%$!5-ZWJ}h+?C-c
z5F;s01U#gGen3V$4cvkgy?%W>kJ2F$LmEe1BRgpr?Y@%haTJKck-3QCzlLrG^P}&u
zOmc;t(yWX~{^n*KvJ&27F(3rm8OJTUf6yh;TARZMeF)Q-8W2u55lXeMghP@U0(lth
zQIs*?Wh0UrPlB`gAr~7Q9XD<qs!IRbS#nA-&Z#~(v#$v*Tj&Cr?H?NI{!OHXeaKF*
zC>x)$Fz%teC;r2S(F4*$*MyqhyVveLL@ZW@G>>tE9ZZ(BfvDS43&1|gp<t~HPK3s<
zDWfJ$;%7{65Q5!R*=evKl0d*KrXJ{gcYbt;CVk4!N0f`Q@WayhLm{m|N)XI_(HZU^
z0pJ*l4q>SD2~JV@4N9KL9TK#sPQrR*gSLuP-(5{;t7-ER`Mk)MJ-mOJ+-{!lF3z?Z
zkCrRPHZAx*hx!vVG9Bx==xSB2r$2{K&^Sr^;Gz3(7OW_0-tO7@uWqihd8vjCywfW?
zJdC|#5v}nMl=Owm?=u1c{3D4fjlIf#>{JyYbLYli>2Z0!X$Kgc#>pHh)rSuV-EN6+
zq(uOFO$JaPUgq$~)cYxk?_b$neKsqzZy%Q%)vx<CxEa=E>&Nv*Y}@?HfGu0Sw(X4R
zpItBh+vuQpyTlE{QoWwa4*PZ8-i^Pm`Yd6*mDRhBo!nC9q}-Xdck@dJ)9l}K>^Srj
z!<}av5B_MPuixkY@L2!R6GML5xMKE=^*3UBZYuEh>$&OSj(}IvcDSe2Tfu0!uzR0J
z{N2$|ma;#htbRmcX^LGq&<7~hO7k7(4mr=;H6Mm+hj_YzGhf6$Zau8u2AJeIq}I|D
z9@Pj`;vi?#!jVp2>^?nl_5)W3hqv=iy^!VhQ{?@>9*)g~P5>h-+aKaOYCwvUZzHlF
zh{a8IZ`thCS8paC-(Nra@%EVf4-FGS4iC+n)@nuftLGeBdO4)fEVnq^ndDs~d3MG1
z&R(RCr}Hh@<>=7|nU*IKK%1ynn6Bv-6@vZ4^ih~ODV!M(Z40g!{_d*dC5<|asAU~K
zT(;}t)O3{i;AX=fv^8L$nmf;R-tZB2cGr^wc<2xpblrz%YB*1xC~TAfIdcZGutc=-
zggS;4)6brp-~*0sJ_Qw_F=Xxn>k(pq><Rb*6^bn!g;oSmE{gXmSiUIZG1}lJITK|R
zr&!=7-|~}NN@$tNUhyNJ+~yw6fqq5+xZ+*)$|#XV`5$fae2yl^7^D&utxyEEx@UMh
zV<5E4mZ6@ctxx7xHuM0jA585hr6|6W>~p2|(WxNPHuGQSS+0@wBWRm-)0QyPV`CL{
zdU5*wxr<FTqV)~3aK2Yyh=RoA0!Gs?=R}LSN@kLHU3DHmT{DJ^YMoTKv73?1$uE>2
zYa2UZD98e9fA(7PZhLAC(WK<i<@H=pKKYmL-$WFi$ZnwQYXIkh)Qrs^JtaFe9$W3n
zIBgZlCk_;Nkj_f7oM|`9L!3g{3N&K`T%?kHKb3SGa|5-zq8WM9zrFhEPVSNRb`_~K
zzO*3<p7$=j;9q+GtH(dGW>_5U?)dS60&1vmHNurzJ+*HBimV%KQe`P&px<fmmK}W`
zea_KniS`quN1YU#>xA9IWxd$pe^3*@NvohxSG*|FiJc^pI5DIEY}PK=yGLyZBiqg|
zL~LTykX_1k{_Q@R@inG8Gv9TTr<)7D8zwm_wx_L?)#<FPct1ZB!EF$wD_WmsL9H~~
zmIlW|+=mx#`^n?#{-u(41<FLloj*q#EaML58mS;c?>w6<Bp<vw%Co>hgH{!>ptu>%
zMo62VbGqtPasgcV2ehUVru_43k);@WvPWk1&Lua-9iX#)g~p$v;Mq|oqXcpWK(T=f
zIJh<cVMYYNy-T$8x+f>@?Yd7kgtF?Es2>(YW)=jWDYXxBFeq3QAkw)ETi#G5PZHh{
z(QP1%`1*)u(E%CLo+;sZv}X1hp5RbI(7@mX!b%PS7U5%G{!s}=?<U{%Asyr#vOXQ2
zDzPXE(_kC-`BnP{<f_6`B+9MXh#l6(I@jT3@#CU1>t%6$Bz7#1S2pLTpZ1U4Gp3YM
zL4yd?4I|Z!FYY2IC0SKUvZeo$jKdgl&}t?<h9U+SrWM}?3WIuUPM7HUKNRrGcjwlL
z<QJ~HvES)OA5@X-q*5+NFNxD+ARVIyBk<<7{^3^#*xLSZdSVhxp+}>+L6*~P=VR3D
z$dp!MH*az;=Gxrnn6Q(#EZ%nk0=0nG>wZMckZAiOaZIH$B!We*FIUM3LL?PVIVz%v
zZmDcYZx{r%CsYv_^P&3VCzh1S56AUnAF~fJvGXDM`_p@khR8F@=XId4gcs?Q;zUn4
zaUyr^xKb!QZu~HyaF?=8<UkqMyY6!sJQz7YgIxhi!k~on<)gc{yqoN0!tKNA<_4M#
zZZ(Y=@5p|Q5n-vG$B0@pT15T5bNB8Cf9kpJ^mdHopBR$?9F`mgem5ktL6=fFWNw*W
zCs6^PX^jEm$^$m`jWrgAC`?gG=7BlB=2=iDOPWy;7ye|{9s8)phc1M5x3lUMVI!9g
zwYp#xq*sGMKYCYFt(hCd%z(O+DgOq^x6<st(4^ft9n(as4v>Gu()r{_2G^li2v{&}
zT69W8oxjX%+7XY&w_FEJaE)+SNw((#-UeSzvTW`c0`rI=onaHfgCa_XzI-bLOnSy4
zQ1MfikvEY*>oEpK+Qt~1fZ3FV%F`KGH0jLxA@nR!k#&9Gk)oXa^*$%c*OSO)`6AeT
z=H_>h=lFJ$hK(=Mk*j})6jWX_2bw)pM#4A*DAO`Yu4!KZlB0Rn1BkeXYKVOIvJyfS
zSl2D85vC=I%+R4lb<fhYQ9Mf-#Lm&d528gI*rC@Cfl{*%UDRP<+}aek6jsD$T@}G_
zqK*0Qw-&n^9S%3T#X1V<VUj-dOZ14^;x3)E0Z^!Y1+n|@7}&|<i-&hwH-NqITzpk&
zhnmS5uuwEJvrf-Qo48lgRm&E9?$S1UWOVnNbwNlA^JBT6+``&8<Q<23blLL;$UzeH
zQwpL89=-u0e>L7dU}Vpj_h(-HL=K>`qBkN|uU`GOA(rc5VcoLsXWCugpT2bq)vX>{
zsiR>p<b_D|_m86J-{Ta{DLOE8Mx&P9+ZYtW%E|eL>ep|~r!c_Ae`Eao6`9bCr~mO@
zZ_4uI*Q;(l=hHyJm8=@Fy5sBJzg7F)<E=`&?GGz@$!zn-Tv_*bU+oAtcIehmFX#kh
z&lTA%zLk1JTr$whhBH3sw1{+#uiTGga(t9i6kC`fk#LZCr@N9at@NUb`!YeG<ympK
zy2^)L=QD1YBQjELcU+Hw08ot1DQ{48t*$0EOY~^MaC2CAC<6_4UVvac9V(LcN{22m
zlcUA(%F?oWY*K06Q{AF$=7aFbSy;-(7_Wn?s<OmS3G5)q$t>a&?5Lc7jwM&?F6(8-
zX>nwoQkt;wemVt*yT6Su)J4PWX4Y=klO60i7^WUI{qgo#tEcK<i?w_3WT&^^l1SY+
zGy|#1u}f|yF2n4@(p^X@^1uIXZgemgQvV#8{uj%-Vcuq&mz(+(vxPRl8x+;X)T5d0
z+-9o3|4;vT;g@jBD1rkv@yM%ng<Ix0nQbHd60TC6jg_5i4<%ucyARCk7PksbY>EXh
z(!LPoQZHnosnlplS`-5cE#wS&_^|7I6q%CqODM*JEv@*`0}i3~;AHKOX0Kfh_Js;8
z`&JkJg$0>)UwrH4I>2SzgjKz|g{FAh-spY$F}!Qb0#{CQi=Om%!s?0--W@2$nB@`b
zpC0BAO0(UOaY0UNsvTld(q6^$NxzBV^7`sum#lp}7R~}KN6n*b(Bz=;doQ{jcq{oU
z%DMuD9|Lm~4ITW936<ZuM{K5%+78TlctVO<{HWjlr%{pV;gip%Vz{~~6APCMxhNjD
zWCM~>!y-%Hy?gd(uqoaOc0g9-fY*H=Y~cbmU+#mhESW0Mpy3@PkEyE)-yN}^LQf_?
z35Egm?s2l?la2f7T4$yj2r3~yQu)eKWh&c|2Q_W|I|NJX=BU@^7EY!&D<^(8bj~uP
zX|g7wCp3D4w|&(J`QRI7hEG1y^f`gvpiMi*G2!oTxIH+RWb#E`p$!ngh_;p*Ey*^O
zn2NU$9{};h99&NC!<}chzL-T=+<OdiB@{=&XI!NpfH{??M1l^*SEB16Xb8Ac)xxYA
z#bc)(oIc}I6J=wOK>~ugBx?vA23Eo=<TDAgibT2sQ1oFqMOFHuc)i)d?;#Ul*uRwP
zK{cv0EtK@2zcu6nXC4)B;fXTTsTk-3%vI203Q}@zh;?eD1nw|98pI$`?uc*)x(PuF
z8iJB^TVNYNi6#*;n$(g-AdLu8L3yqix|N;NUFQO+$3E1wQctlDr79+d00wD9!ji4~
zCqf}5v*pQD#p5guFV5Ep;U+{C(IYiHL3FwM)t^nkixhxB!5O`$J9ZdhX=xdE<GZv|
zqkg&X4(PnL%gn8-+q;CwwoN@!cCI@W82exvp}xs93!>~HuQA7nBjGr>cJFFW=Vm5+
z%7Q%T&(O*oqPKs;pv-Mf%A}cGnlv%5z#R(eQqmOSd3%a3?qavE{!Yc0M9R04`1unh
zmP^Nj-db{6h!BqFv%d35Bt5>og{4kBMe$Vj8E@aBAWfM#8p5Hv>>k{DUhXEp{Uz4j
ziJ}3q$yjk3K^QcCGWut2;$+cl+a(2R3x(||w<rnJ=d2yak*#x{!}T7SjcrI;LOCow
z4X(ck*0UQpMB2em4T5Vlfd|pL)ChtZK)p1Ki?>tB%aJp;-l*0m$L{`R;@&AdN{9>3
zKh>!F;2)C?zsEDqI!r|FMMPeARnWR!$|u6++mDsJ_kN3>^M^|js2rt@8-0i6A8;`N
z)JKANAtGs;`O;d@Errk>g5OETl`vb`jQA?Md=-zTlsJghDQ=z@J|2DNI_?Z;%{Fhm
zaU`pl#BYc(iD$8xJ32U|db%H8xf{-KTd&HzL1k-!g=ns5tlIIc^7Ll=P6f<s*W|mN
z$Ozd@mds>!WE(^K=qaCLfHOBpHN-+ID&o9MCXThUftcAlo{=BW!11W7VdU~mnyzCH
zvCcik$*^jH94Nk7mNJW~jswdZH8&z(O?RbhKgW$fT1FlmTVh$=_eleQPa5h#3q!+?
zY$8^eHvIMv%gf0dK6~Y5no%i7z@w0jIW2nP3ZQJxtO*hUv^YF6+(7{^d^<ZTaF^xw
zhmF?O)}?Dpvu+O0$!O*5fsWo7QP_861R^U1sIgLeE*a<1Y2AW~Z$qQ4V`w00s)c%l
z(0>fPyU!y-ChjC!DvOw(s=1p+AY~qs;yr#E1QAF}-nFy`x0anRrnq@$3mcpC;PCm~
z8aXb=iC^|@+2qCUrH>EdN1Z&f54!;ArP!opXU7jcAwZOKiWN8awZGUp{(X=EwDib;
z776w@ElwUzEX(iVnQ-dVzPfq2fMkhvAXmLn*BuVOAGP(?2<3fR7FLuVWH}qh`@NWv
za_VRf%|^FU4ltLw%h!(X*fca{Nr;Ja`*2xko6+H4uN!&HZ?bFFT-t(lHV%EPiMXQ3
zJW1AO)@q(^Lgh59lPDleCPP4gISb%}bKifpgQevUbFN*P_^_TM!4>R6rE5<-`|Pte
zQ7PB=*ALqtzTol6Wu0T`aKG?kt97Suyg&Xl4B~a0u&|T6|0SyQPuSFq`SZT_LDR98
z#B-SDE7pZ19I9%+;CNs0a6cf1Q9p9w!Uc;ZdDeOOZlgMNju>4>;+Q2jr4u1eaTHal
z0^SnFow^?g&v<`83x#6MnK$q;wiFTOlvxK>#aOyrwD?Kp0sU~Ls!)%wFTYDxC(~<}
zz1}M;=V5%OS=ak9o7SB=vwg)YR!*{MQ6F539p_>@?{nwhfn1sA9>mOid^PMDoVd~)
z3ijNa@9w)ewI*wlAuc^0T>g?y-_zzkNQ8AK2lCt4kn_*&XR;|uL*;pt1%%VAQVR9w
z*NE9{zPuOAf~Xz?)r{xbacr{2K*{`e?K+p`I&|n;-yM6)Acua$yk$nsGc^q57DC%z
zKNmp!z<0;C;*83=<r_zhYYc?@>=yRfVIF<js~fvKo^ksPPP2!AP=+r0LXMYE&MTp2
zs)-Z^E{q83=BUHdXj_Dovq{|}p?JjW3+@exIzf9M)(+Pbo4gcCiyO8G!7OxB*3p{z
zm}}{?$M>rGuc5IczOKM#`aKNcO2FTR?)yB3k?VNKsC$$$fuUX7_J0LlO`EzTmBU?9
zJT`w`(iH2bMhRGxXG%vg#M${$gJmn4J(hkry@GX_q4LXMg%ujGXF`JV#icFgAW>tG
z)K!Q(Wr7q}oX(l&6s47Z$R`@_;`mPMqQ)Mp=MppHBXFgyxV7_lRY#c1HpVffgcEeW
z)_zs<>13ur*@`UE)E%^icudZY35qaRx1_&iE(*Aq#?wh^$$|im9}*NX#3<`DC#cMu
z0G}EXQWtH;4V0zZetx|(Qi~%DnaoTJHe1b45I#XK20WB7cK2jV)nV7e`y3o!7gy8}
zz-nmG?4w7IPS|_VqHq-3ChC#PdpdAysY=7{ISUWqaA~kY*{Bz*eZXwX4tYhsgSdo5
zL(hw^5Hd-V=Y02);H<gJ-Cees*MU{LORVNwA%}~-u{SR*a#KRpKcQd)GhCbtI{y{9
z)tKzk{4d=f10nxSr*gHmJFU))8Mli&MM<h{wy^)$XH3b=$wN$Q(`;~G)0u1bVT<1+
z?w-GilK$+E?IZWsb%zhEn!P{1pGSb1S>ylw8OI*Wqh@f5sBL3P)ySH*ZTOF-cJt3Z
ze)n@72KRM27&ky4rhgVX9j{m01~MY{|GHEAeGk|W<dnX_#h02xV>T&ENpug}ZyWv|
z=z1`X8WRS56LPBn^nl~oqe4dZ?3~}z)Hrs*Z#{%TM}+-Qo#oyFJW^7SglufzH$kVO
zT&68+!@uQ^u-jX0%7ABJ28u&y%9b9G71$W#b4oW)Go4-w|80L(^!{{Pj#E@YpbdjP
z0{))e#q>?=45FAao<E5FJZevV<0FktcmF>%Ux&O)F2J5EA(h&mWB&V#Q&%4wI|TN3
z(KYEQ0Cya)$diwVf@r&#H-d=sF5tH%uF8!nul))62N5_k?_-0TE6t9~NK$zyUX^MN
zta0vXca^S`#cUH3-d`gOuf|}~wr%jO^>k{Q;S)<|CV{HvKGB)s^}A`ETQk9EfaZso
zD@F>H3I8-aVVC1;7|L8S1T-pFX^V*s9LO7`f>VTgZeU{<bqc;OC2CAe7gbom_2NqL
zp%&de{`1eju*cT9a;Tt(T#0gwUd%YSL(HE?SF;j@990&10>*ld0gj~)?Q%OW2z@L$
zI7JF=I|Hq~+Pp)Rpmky)P){*@DpG+3<hn|CNvDOuK4~~5`7pZCjwIE=*5FHWm8V`U
z0Qwx<Bwd=1s`_|e=$AROayoodL4e_QTGdbpKXFZzCv&j!*3i)i`vK{@1&qV-ZoxrL
zJEysrbY4>D%gKjM7~{?io7@Ar8;SS-O1msA7|6AZjf%Zcpe-pvqwZ$?%U*9me4u5A
zsq5&TH0h8Z!-R{QBz=`#-WQU#26aDGw;Q~Xyct}EFaU*6dQX}t2bv$90=Q5=1*6ny
z3Qs0VneH%Ve%Dc+?)aIPC`v$J>sxPq+^EOB@1^%fr!$2t<<5`hhumFv*xv-m44y)h
zRt_X?J^F}QO^!VPdtoIUlUNIdy<(vVM?jB<DEhgb3V}HKIp?^2EsukZ+Eo^0o#PPv
znz1Pa7anrp7asPC$EjvY58SGc#=@wUB9>QrJgQ3|oRfL$foG>M^+uu2bbPwRoW;Sk
zDpcgdcmSzWa%T)g`X_G#HONi)d^EJ@XllfJNXf#QMKE{6+MTfST01wYDf;}q@l%UF
zErC#539gU8-5wBGdF@VVBapTa#E>cYP4CNYWZ)D@Bx){<l1Y=F$_`Tnhteh~26$K-
z?^f2`Q5awakD~(hCxinQJCBBL6$O2dr{h=V4ok_t7j6O#%mSSDPi?#H{LNeV#5amN
zM{kCoax&Rl4RZs7f(*u%evf7@DWig5SDZ}0-O#VoCVFiqG)c(}#J@M1s#=5LmBOM4
zxm2)@6*gc<WJpagBx*z=2$s(%)|}GODpvm`z7yc4A(8eQ6U*r7b4l-`Z4yjK$<yu8
z8o8f-*Ph+3Y0v!Aw{Ll?vv<6@Ts(bF;4l*TL`Cks1f5ZsX778ZuD@Dh_FI2NJ2m&!
z+)v#m<6ffDZrwS$`qK@RJ?D81kZFsuuOrUf4bGol?$D}^d{}5p>%v1a`@z#Fd02VM
zFJY}tZ1j8rnW?r$ac}sQy57^|G+eC>Eq#A58kGmbo)Z@~#?&#{SJ)`4h9o$2HtBC(
zvYPb%8W7ZL=OzNxvN$9`g!J(2I_;rF2KsHEk<e}X5Y$U)-ta{F)J_%qn>6;Y8uA*y
zTETw^A}_={-nU60JGonw|MZJjLLz~;jqC5vIJ~kc7p!;8$TMVyO^L|wzA2<?`}8R9
zs7@c$o}Ru*oMFZ9|Mk#nh50HP4(kB_C~877B6d@WPM1te^e<{jf}i=xoO(`q@nbTy
z{mgyQ2>R0AkPpDN=)@(z>Pq;FeSWe6a+|TpjvDiM8b`Y!kp#Tv>HoT!b3eQ){C&5p
zzAUM(^W#?*tcoc7(RrnNY0CHpISc&lQ*wv-XV394L6jdZLt|jaDav=*N0*>bX!Yf_
zGumjKLkO1}nds-pw(9K$XO-5x^SkfvO0;&a%)JxgFxNGF>$nKdY4fr<jfy;gPlcKj
zCyS`qnxr$~_=wFond}?>>KFcYaaxLHM1p;(&Gg0jse}Df_NL71Y5#CQ+OwvUH8}>v
z9{l6NsC<bGwfca>q$rkE9c>~B5S%_PL>Nq6R=K94MsbkMNM7Uk>X+v`;w;a(m02@g
z8W8T{)J_3Wbe^`e>JRJ7EI?#Z`hAL3IC`6n&OWBzq|VoT?eE*TCx(Yz`E=u;3BE~D
zU({IgxSfMrDQ=w7LUUXi7NK3C7LfDP{Dheov;&TqR21|s%MV}(5;x8%*zlUc(e)pm
zQH&--N%HHK)ATt!J4!l`M{~z)xlG}Oy6&SvRxpKZO`O;+shwzy3`1!ITf2-H6UVIp
zG1+lzaO^C<X?C8)X;F5$Hty-+AzN(D?y!l^8~IqTZ@m?i5}8LT+g1qCx=RjIMz;Y%
zF{;n>si;vH^VnhO`2J{DBC?=T1VkOp+MVOGap0z&nPXCNFsYEV{4VT4x25BtRKa2s
zfVUKWL_=4!;FS)p$x!|-i;v}8$;92!;^T1~nXDg2z_uif=Am(|=uRNF^jpdZU$UFG
z4u=&n7*PjonC|)vzx?K{LpJvdyhFS<bUqrs^MC#n+_g91RaTlS8609+&itC{W=i5G
z2+IN4JpYo1o?1Fp4kAeafnKDw;>b?k*0&MYj7ABBv%0t=bC>_U@*$*%Xx_weD|&3_
zR_#QOo=+;V8eu$28qPy8;bpS}$;8?ZCM{>1N17$Iy4W3j!iw&j08EZzGIfmY7xHHt
zH536yi`3=nCR@zCv_<t1XK#9}XQ-6fe9w9%LH|=CVOLb)*(2q^nYCV_Zaeeu2A}!e
zPWj=oc_@iXff{}pyL%_;PBJe}oVZ;_1{--sb#hhOe`?hlp1^=IidiG)@uW$kK(v<v
z*(qnJBd9>n<0NeEaHnbgXRSu;`2^8RB4(++XwWCq5XuxFauHQ}K-$aysc$`cn^WAL
zmJ%_7<b?vVrB$YJm6{$t7Vf$)2r(t_73-25^jA`aXaX`fIpy#u{~;*_2Va}krvbY|
z%EJNK{r}KCB6r9pFtbJmafY)whz(*`-mWl~0V7QuDMN>i66M<1pB0PaN^s_!35`a{
zgo>G(jn~Zic_x=boLTy`Wxq-TWppW1y~gR-&_g+Q*Jc&B?BPu-k&x5H{?I6jfsk9}
zr|ffV=vytDcX;cqtv?+uNEtHBx}Wf2;jsa^_BjvxrBn>awlMVq4_tt{X%%N2y|A^8
zJT}Y!e@TP=zPb9-cMfsPsH1HkttKdE|9)Guh4pruOwoV(vau!!v->ZX?)yc#d>j<@
z!h%R%Jmb6n{Nn%b8NlEF^#7`_bj=7dvo~2Z%{15#v`Y?Me74Pdau`Sz3gM9g)dHgw
zygwma@(}&e>4~95Ll;{7w_W2G+194htfo)g2kmG`>{fE{Xo%zHCMIlP^C@f|ZEB31
zG=f%+Pd)Pe?SJw;<dz2Edyjw;H=b^!WIw>5vGGxSAkML9E+LwtP9UV!!l77D4!8sY
zC)%@oY1@#l&g_cpdKGOo11k#c?y}78VQ~m$ok)C(lZ8<e2i#rVsipz(Th)$_2Lj4t
zr?F#FZqn}m(AKq9lmtH6m4H<M`ShS|NDbKfX>JEAs}$6<p#;f`MI_)DFKh6OW2K;>
z*39L}<iHlW$52@fSdqdi=XT1xGdt&**3Qm7pj5U-zC6VI)akuQ<iqr&Y;;<GkzIJz
zWtXMSKrUO{G4($aSpo5XEy^E#!1<x=Ro~OlE409xDyY~FT3~52lmsQaH`;_VIqLNO
z9dXxu`Pq{v&$DBw%VA+Tpo2{XI>yp7PSPm$K$)bahbJ~thX#IX;;v#DY`Wgn$uaZc
zTt=9(%S;A&V?HPnl^%hxl~mAURAeN1_CzqD|3RuUI!GGB4NU!z)JA?xgau2kh2C91
zb9QoQQH`lCIpWrh^Ma-}Hxx9&2K;*_?T{Z4ahgOzd#GuQ(@94u(RtQ`B*%tht^hTp
z)SjRssA)KUnA7@um;F=M=$N4(7aLg92o$)2*~#9#lb<M~0Qr}ed8aIrDN5n1{?Nn?
z;Y&bqo4btI{fJ2~${GeR+(c6L%{kNH;iMUIgZwR<-O~H!n;hLF6Mv-3R@suG$We|n
z-T@M&@oTTmJ9@VlJr|zjeYG8irF(Z`4R9v}(wj2T#ZXqU2E!c(c`zU`U7-@4R@{}H
zA6so2QRl%NU_L(fn?9J<vSx{1o_pqQ%cRmqPntby4}_Q)T$E-xuGL#_btEcbNfq8u
zL!t)KpTnTJg_kn!7vbE+-d8w0k%3J{m0UK%(Fjf1k|~7p*=bGIjVrJ(c-))K`pewb
z!7Mv5TCBqDl6tQ1Z0gMluY{OW3{Iix6Zn7!Ut@)NB$v#_cRsq&+=R39NWYUt5>)=e
zQ5RF!B{;t=;+9WuOsDJW9b;*zYq<7t=xWXhrLs{D3?O<Qx-)NH@*f|+I5c|7&ivWO
zPxr1Nv>%-bFP~XpwP|+0zPU%t%x+~|BP4{7@Gkpc3E;<>Z&>~V|NZ)3-{?P-_yUCy
zDhp})!P|?0iN};QqeM$;PV+4E7o6E8^<9R0O56nH5#n{k=ibdB;!Zh;8(x{#73z}$
z2`nW&gO*0~L!k*mz;)Z7pa=`ok6*hs|8KiK?c1$-(Ps}GnnaXCn_J|?`!vhGIq=J^
z*G*c0M@_yvmUpqAkvzJi-Hpft-vM4Yk~bVne)hN5@{8ULDPKSHt_eRn5(Fz2SO1g9
z94r}WXl}WE8ZyV64m(Bku}^A@&J)|_Oo&?o<ZYN$&zRu<+;_w-M`ZQsC+x^PnC`!E
z%h`-(==StM<c9$%Zua0Wmd=X=x7PGz>&usZI&RTym(a-VH#(GrhM_`}FGf295m4yK
zd-65T=6rW4Fr>$%OU{osyZ6}5IDhN`IJ6zn*%v@0Cxb4%ZKzTFP><V|mskunk!05(
z$sA%AkBS`Ve|=W#f)opz8&0{a?;~(HkQile@0lBUD)le_rW<SDyDFsn<8N@Q+nicZ
zdtrt%mEZW?!}WiAFDh-~-*;d|Y2M5$TT;Tv=>YSK(amN|0K;pTT{GyOVl#L~hl?~_
z2KyH_Sr#~)>4kh!<5022DC(klIZB&|dDA9r{9sx{LG|6bGG*}dTwF(bJgF_0c#$_l
z)P^k9#&Krde#e)caDFV?Zn)$&5ZZHbq##5fez|du$h#CeZdLKx@{+o$fx&9lJe|>H
z$%u9luS|i8ZSTI0+TL7q2-;Un|2ceZHU|UCn!;`1m54bxYhv9(e_Jeo<4w<Fde>+I
zgrW8zjlHvJotVCio|AaBLkm#A83R$9;M@D<%6i(d5$&32*XE2bE4x$af=ak}*H8EA
z1mn;tEsnnjy^Z#3LX?xzTg9}L9cG7+(z^Z8t9LffVt;J71;u`nj#S{VJPb|bBTQ~x
zJTN8m+lSEZiu*$kDNR7b1$@N*_zO7Ole1hm<St#e#H#7SW=A$f)s&#OMo|<i9gfFc
z4U<y#I0rPkr)H=NTX0R6WzGqyE{h#U(h=5X&0s+4N719Oa|ou^=BHFlUrJih@7k<E
z-EV&dz@ZsFJUPKEY;lt<+<X`reorwM`0o!$^#i$gig+vL52ex&U{Ec*w@1PyH&Xp1
z%DFhz0W6g*W((UeXMO^rYz`&s{ZT)sOiOu$%JB<YwH;I#$_+-^UX20=_RR)RdZ73;
zFcfP9;SjB^(CXoL*z%5=h>hfbph~yv_2|}@6jXd#6e1eQayTZG&$?jb7o!SD?*bFi
z=n51l?zgWVla_$dTdOb~P;A&V2NYYt;LN^=!2&@2H06r%b%KeJy`X=l<e54ABoagE
z&*_Z&`#rk$x}@jKEhykj7gi6E=j=KND*&083Zl}eGcf*PtU`@P;wkC}WNz*Al>3+}
z$a)K;B}_t-28Icow5c76vJQ=ucvCe33K*SzICX}WvZNY1+^Y53ANdg}+q9dHlu?bC
zc4uN&%8X=*5L(n{ZdgAw;xtZk^Iw2DKH!}^`rS78)!47mHe4y7Rb>0`7rUqQ3riWm
zVrGm9%BN|HoUtCKX1xj^8PBBLfIvQ3mZA1%2~-bHnEhCXp5J{^a(7cb{HQ(F<@EMX
z#EhBxc)skP5mTmW_`f32Qff<}h1q|vD}g|7Ce&`PgKZxyVP~09LuGkBS^ZS~%M)#_
zi{o-;HyrYd)uxx@t)k24o|xXyY+*(tq)3|wUiMqaQW-olebHe@NVE?Z2^{}$(eSP{
z3}(EC-gfiH8}I+at!A2n*%9lQ6q8BRY+?Qwae_O~g}fe}mH2S+oT>9{bFs4QT>A1m
zsmWf?``;+2h^IzdIC5!Y-lyiD|AsJB2fVOga%zt~v|#$@NNdyW{qqY{2>W+r`h{b0
z8GN!{v@rYK|NJV`sWjuSufEy_%=!+k!pn5W7K}%sBZ_j#x7MLXkD;PI#bzVILBw_J
zs3~qOVtU**){Lt>QgUjsPjP(E-3AAO_C8^~4P`rf3%t0(ae9LXY^*)&L#R6yUF7%R
z=<9C5iSB!y!X>IKeIpj)j#orE_3ne-D+S@9x&{e1zcF388B;iaHM8yL4jy`WDsg?v
z!8MtxJIdwlkdtmv76|6h;f53qo$*v18~VWIp<mWA_qeAwpJ9KJt};lzf)b|7ZvSiz
zlUxU9On4&R)<h0}=&$smnsMMvjdF^3_Ce)am-}o;4fDBfRjsRPW~QpJ8nW?EJGSP_
zA*9yP&A9qidIrqCjsT)Iud-ng0zz*Bg;NgBn~mAy6Re=gK(0Qgy*oI^Y*y+xi<R*;
zazMPJ5jG9e-mGc6fI1jzE!_%qn+#A`W4yYwM^G>OD6Arc8d1zOLk0o^udnrljOwc?
zc8JlWa+b^p*f8SOgFod1CojsiyHRzpEZx1V(x3OTKu^ipwq%^3J;lo+(MFL#dz!?X
zW2Y+YIB7yjGq;u|NnglRP#&W>FFU@hI<E92uTduhc}3kiik(a{56(j66L3~m?50v&
z&0c!r%!Y!$?dzY|tLk<mD1m%R_~>XQRjS}3vui8zU!^(<sb&KTq9Q#%A^bl`F^}#h
z@j)8JKeB8MfAm0`sXh`G2hE@bESc^J<`|}a?}Vrc@ci}vQi_?k{#4c`uu#SL8_gxb
zfd4(){qE93emob-vekqK$WX90TZQey+JQED3S`die5tu<N$%VyUjr4Wv~K28Xy_VN
zlnexIgwYsE&=UNs5*<)~j*|G@JP&|#&+F{rQt2oOObYd*-4XojVVC`kW4M3pOEydk
z2%%?^4w7+&O1V+Zj?*R$eofIoJ7g1!(EJ?4DGDYMutyyf`oEex^ROJ(Z;!t=%Ctk9
zWXQCI6gwe9#*Hn>6wxFpWk?#(ASv1;nJT3e*+?QoqEsqFiKHYYq7l(R14V<*=Z@d+
zoa;L0k8_>B&U;<kHuS#l^W4vUuk~Hu^<8UmSAumT<DNYkr^iQ=8e9o`Fwn86-F%@i
z6LMmtt-zVtG!>w9xI*NJMU522qd>tePM`iK9cC^S;KSyN%4oKEOCB*ajKE?%;u92=
z;v$6`CHxj8$%GOGJL(pC)LgW4B{M6gD$VynBvevTQu&9W#{WWgdF83gy`Lcw=lTDT
zuDiI6WKUa7t>?WOfh>eDb6Qhn6ftw+F!YYbKEuS6bLPWE56Y*BeoK|LtM4$F0*zJk
zAkj}4cR<q+MsIY)lgA8Yf@!}fYQ|xBD(e=AQNfDy#k^(m(Fvkb69vq2k_CeNxUR5g
z*ulQz{&H>14yoKm>tym)D2MbZn-JZRTJnH~z)&G{1=q@ycB2`ZPjK{x^B-Gc4G`?0
zLlg-^x@Kk+Y=-=2O8EzdcrU?56U%<ux_NHHuM>rDY<n5e9v%%q$h{HZLKuJ?G_#_O
zQdo~aQ0Vx4CetyBfusLe!>_;%*>*AzZI_7hrdVZPWen(i(S33BhqQrl<Rapz7|Wsc
zRG0zh;36qJF<fSmOIuz!_$`s2LGcaxs|Y=$&3asepaw)a1JP96)GIr5H<7towtB9i
z=FKj#8g%=n4d8n*ziLu07C9#M>z9JmiFU4+{f{0U2K`H)nb+p6>_f%*1gCc8TecP2
zZ21WtxVP$j(Owee!io>UZP~k4SLcR7&+8IeU7K?kduw}DuX&p~YJAqD1zuKpM-@(w
zeYY}y!#g$U$B~Z%9#6i$#-h$_Ol80cg##bfs~^}<q4Lsu%}(>lJNGtsTKe9j{8B;T
zjK=!u1r-W@pC7eskG*MBkZ>~ddEGb1L|N<ZnGK~*@45l~%t$tqiaA2DNRM8<<}6v#
zgLVuSEmoU0$xqup9_e)%w!v}xb_Nk=LfKORIT={oUe)i9kxG$WiYREW+0Ms2x-R7g
znwE?A_x2`{(vEX;H9&Rm-8rLf0|xis(AT>4%emLlNms(@WRnTk!q!}!7h^pL)@#`C
z;RX?w&J#1s8nz4{F`|8d%$lZz)5*24EWC01W;<)8JIWlhwpZ`_4<C9`b`Gn=TZc@U
zavb*T(3Ql-E^c%T_oc;c!0T~DZQ#{Rc=o_4+eRfd*LKAer=Ogbe^iq`p3+^qZ&SYL
z;}aRPCBF8T++0Q1k%{N4NW_O8x|f#L#nRG}WS{nPT;1PENi(1t9G#rTBGD$7QvLaJ
z^6S?LdL=EJPi9`)Z@*My@?=EKw)u(prIgFzDzo1%+r9Fm!J@icU4`8MR`FUs4uK&J
zj{{|1v-O7#8}>H^c}E08vMFo>mej`1&Xitwx9AE$e@gkUtCb{f2+7(Zrrjl5_KHPf
zQn$r6Gi9ADnLMZ<+<MC2RysPeCX>H8A!ZPX+R8@E6RFSJ$c(pIaTJCrD)v-RSjdrm
z+}KBPi?#03%+%amIn?!JmO9h9Z>62lsAC(eA^h52ntSN=#JdlZ7!Fz$bM4w`q8ROv
z=?ZQnqyldk8XArxUKnF^bEc8e08V=`vCG8#(3jvOiW%4=()2e~Qg_hO>Ox{=B(kDJ
zY2~KX!IT%SH8+n$sNv<Zf8(oq$}#lv3tPWCcr`>A+~_hgVrjQM@#`sTve~*-#sAfw
zyJgMJy}#{XB=9VIdv)pV1Fz>LV)~(b0dAQJsXoq&I~%j)ZLjbiP!}VVlsvilD9%z;
zG+hqFp+8k^a-)#=qT^sz8<$ElX?a8@sT0Zj!lW6v+jAA@6nQ4?RGF&0UthcPexGNI
zdMm>HoMW7>H$CP807BK+rsArqRh0#V-Ph=68D)tw;BOC)Qrcon*@Wn+3tC+$@<|ZJ
z`mLCy#`>zRZg)&nEtNmj!M=`dihKV;x?@LMMtVd=!^Zq3CwCvqx7<4B_!3X4o;{U%
zx1@}2+u07L=H@BXOU&mCBC03z<IWa5a{PE2;h(B}H@?+Sd*^hzG*H3g27HOQbg6T<
zZYsHW%IVt2$9uB(t7!oIlo5c)Q&Q32Kgh_iz^8ZU*ztAt1pD%5p4h8us!rK$U6Fw=
zqM9@>ZksaP2V8~R4jg@oR{u$pCOx5gf}W-aNY66)RzB%_qcUAhlkeZ(OBW6u;?QNS
z1ku^-JE*8f1n3N==!hR)hb0hKg5c?@<?}Jjv`NR2EcAHc_x78fp<(HpH*@~{^B3^q
zo#)Sw!HQ*b9hm&ZR36k+SDykgz*(IH25g~dhItm^;-aIqMz4@=9JBA)l7qi&U$L#z
zPeYJ`jKl8zlObTY1zWd{2i)C^i>pP`F%EAmIApH|eKsv%VEWgh_LSS((w89b6S-p(
zx(}tF&^QJ$2!MH?d#S%@!u~!xEHlCA_<TplNuXa-QWcvDZUxeH2>D;2x>Y}9E~hCZ
zPx%NhH-><Ju$r0=i%>^z8QlI|(wu>6=3VHFHLjwSDN5eFlkDRi;K$XN81aY(OtO&x
z#SkU-ow9W#vCJCC8%*z2qs(={RyJcK%sv%xcR3^NI?2ckpFBAb>`}`%$oB2)>ER*$
zDXDcjaQX~XM?=v=EvJ#fE{6K}#-I0D7rgpmou@d;n%)oyg*h#l=~4;}{-|@od`OyM
zz7wrxAQVgd-dLe=sO;0H!&&Q9B3s0d$dyrtur1$tzoWBr9i=+2A_v_@-($Xg8cF(P
zmtJesG&Gb^gdCHX3xQ8!U9ZF?)M>heZCc93%H<0-9a*i(1?3$Yp)Q+sYK|85_O`3B
zes$%-Y~-)Jyba!9Vbg;qcE<!!t74Q>yziO(l(h@z{Cr(_1y%hZEv*D7aQmf!lCt;H
zr!xO}($&~CXs>q;R#NIbZ{9o-J1cXBcs?7cRlQ#`cG8V45#>ED{H?X@hcR0o!M|^5
z{Jeq1)<;Gx^GR0ecSXm!XN}Z@^K`@Oey#kxBs62Ck6u%<R>+pYR?04-wG0=T#@BS?
zu`SJ6v+tSSsP#Yh8voOo{a>8WfBRFo)RnfSpKv7m);7P#uPJC66=W#x?BgNXTWz1x
z5PP_`a+FJkV$7|}X>Xz*wEQ*0dg#y_hqcE4xMuO|vn4OvJe}#RGj1n;IFRMPEj&Ke
zFlWfVeMir)81A@b%a#{6ZGXoI8O7Rlgq{Vjp-CHup3_fl#oYUY#dq2Lkx*^oenDo|
zo5bS6o2zv__}d%WmljkT$J-wS*^u9$II|}ta5LrAZ#lnImblaT=;)L?1_hC$)UZop
zHJoRKEI;2bvfk5gO5x0xE6R7U4Bs}c+C6dT(B0S{dT`CmJh{hOk5k0O4^~(APaWud
zz|GCgnWBugOVN38LG8ntd#-KIIPKgDMMn<8V@^KuLND-xiM4emMcpY^*WA;K+p2`s
z{i@*b#NQ!e|ABw5f9^h{XV371pBsYs^yPibWKLWz@1bQi-?@QXn=?xPl2$dR^fx!O
zb$u)^u1=aR91l^s;)DskSFU{S{IB(B7|}iCkgmdAmvFOr(;UQ|g*$E3-R(dJV9Cwc
zSp7}3x;4j921DH*dYDstNY{HTUpV4m|1W$Yk5Fku*cn~bihH`MVarF1t%_S-zN5YJ
zHg)kq-ENG!<4ixDQ#N4{inK6nH<*f@lLvKsj^$wQ67gc*7rX+s0?8<roTvo1wX;(T
zic&+V6llD%zBwktMQOwpoiZt#ZC58IB!=YftcYu8XP2?Eyjk~~^T$UD&I_+xENL*(
z-C1aTqUm{szTXswV^YeG6BAdv7F=k|{F<JgJ}RhbXOq*#v`<a5ySuEZ|I%{3-Y}`m
zuf?IioBO@i?-B7K--F%D>s)5IezpnCaH(kOa=v-pjL-gMJ7=^+%+5&+yErDPu-)ZV
z`G$XHrUfPuSkO8yPRFcA&+sZ1ssZ9tlK0ZU*q^3D`bo}XSIkYep$iNxJEgs>5TXob
z4yy%%$rw-LIcMK#80WO(fxYVL+^}+6hoOGy&DlK<9XiB8-M3t|;)2X6wY~@Mj~Z05
z*{r&@Htu2*y2Kx^v-Rh7H5^X_2ZKb4Fcx_)Ui7wP5}1_Mn2;qs&2;B<qw-)VfI=>x
zLdLK8h2QN{zy5sph5mH$7u(`&Zn$n|W1KR*8Xsn4yxS!`aAfYzZ?)zU61zJx?_by_
z4-sm5A?M$Sa;T1vf$nK$+)2BEFbp<PkJ;fAQvZfc?7~_9y0|%S>{*KEnVH?)ME89?
z!_Aw=XxLZM6;~?srGnvqo*J>gyz{w;0*Ib08dnK*5{(fy#cdwn?;8^c=x#SyMWq|X
zj8RQM=KI)+2M-^{Rz#*bhp=uxUYh#)a9n1((n~RePvs|W9sp?h+SHWIz!fS=X92TN
znQEV$8mnh_<NeGieE$<*LVUl9%^#V7+IA7;>(c+p)>Zg3eDsD-##j(d#f#fI8@rJp
zAv^SlpVnNk;9hpNFHnY@smkyb3$&{J*G0r9EdF`Z7y1aBIpw0?zY&-Be!HaMjtBqx
zo$$$w>MI849VAg1eB#6hCZrP4BPH%&Xqa!>fY`?uEm9pYU(K<)sEb$G&N>3Ip2Qd=
zzcy?+9cno=#kI@79{!s|e_!9(CMJWaI+TF7YZz*B#K-3@nEoFB&Bj5#h}|-RkbGi%
z%&Ain_d7F%$nS4s%U;t`0p|I}zlH01M6@|TR`9{={*&#7@mbvlP7T0RvHseC1e#;v
zbKMC74-@-kG3b1A!7crq;{C6l<FI}#ThE@p^nmQqySJF*te3ET0;%-zmJT*HCa4h)
z^qa7sO~T$<HrIQ9r#AoFX5ELzXliNwQbc_d)W*5wo#Z#vB%<BI(VKIU)mX-#OPYL^
zYFG(Ksvx>~cSy(}Dfe9n-j3y!;jo&C&`baE7<d<)j<Pnfxpsko`ri>LX$Pqv<bgr3
ztwP4}03ZvrQ?Hl^x~Z(HdV$jE#TU|u{5~=!MxY4`D?lu2V>fKt5no@k+<GS4d!WLu
zacymIMW5dX-yyn|*r4|?9eYy4;Sj1ZZy%rCcT9@NL7yY@GHJ?`r!3KTc(B)B8q?0G
zSRQ#fw{0uLDpPaROgDfqbhPIt`oieF<JR9H<QQKii49@vX3rRZ{hqk%3mT*i7&<Y@
zxn2cXtm2q4hX@!<Td2gidgDgQ%a?ZBKR`g<OHK7){G1v+XyC@?YigPRTyLS(Pie$7
z2a$p2@$BN_pqNS!t7`{O@}`lF4NiVpA+eN<`<>e$-t_yL+S<t`swS%Pc{gp(=o?~f
zQKE(67QnRumqly1bNpe>Uw=J?2S6JA5Hubz!*x1n|BV<x7in9tv&o90wAnC_0U8ct
zfTLmP`77Dbw6E6F-hRy7#auUcf@am}4#MjnT|m4GbWO?0@dGsZ5jU}pe=b@igIRn>
zs58-b_(Bm)O=VC9+waVU*_U<Zwmx~{nabADdHv$$OW8q#7K1PIqSke%N_`mR@x{yt
zSW6OU!1NsfD26l+jbq#lHE3z{SqhwJJrD-xcaV^Vh6XD`Q*PBS3HDX}MA(p(<qhj<
zhd5odh9H7SCJdgxcyVfBVd2)p1hf*8J!(PY)YawO$h7q6+jm25Js_}sHB8F!%V=Yc
zQv6X=T6&Mz%JcBybr^P|TaMl=lg^SQw<-8uXDd4Vbm-8*hTu~OcgZ2&re@%3WaL;y
z+eh(<*KP)}I;VcD&LjAbz{y*NOGqrKtWFGuVYA`lAVqHj_}&*JXh=|tZUqn+S?h@n
zbGDY^3Okye2>*gm$lc4UD<}DYf`&hG+oc=<LEp83P<hf^rf{5k`uc&F5tC+LSv>>S
zke(*1Vs!HeicNxxOxlH+#%(6H!d6qEB8BjO&b)c;mk`FVvXkuIbYg={{Bn5z{?iyR
zhVGoiU=Zk<Y;2-1$nEaSyK@Qfl1o~LNF=g#qAx(Fu3bw}D@x<lENbYXA~}^TD-D%u
z1{|@_2q162kcft8eDJt&QLieLz2f!VpUDUZxKA^X>b6=ElU?1=bb9clWc&4l(CkiN
zb)O@^4(rjUPbyJ^grTTCXoj#xcW+yKXl}A@T`Ik_?vZ+n-SJJ0+t@)G_rQS3J_NXI
z{k=g!{UJ-QIkwV$Bi$!{F2t2yB{c?*crZh#s4mgt>!eH&S)H>K+?_=1HwjV@v9)2<
zPND&kP)*ml*UrXGgc-~o*v;+=ykGt0%UgCx7&g<Qj<d}S=|0&B{zNSLDvOKD%QyBb
z-JHA9z#&R1T%pa5daE@tD^#+csE`2c>eu@TnwpxOsOpe_ME|rNBOdccL4H3?a&<+N
zafi(%kHID;p9+O3*S)tTv{d+g$eIxZK{Q<$Fkpc8x<0QODV*%@#$gNm*tJ`?4n<_B
z+ZWL?W@fR>Domz?ohX?JGlE$ql73OrOs;VUJfv79Qn0<GrKPjOvb1W^r4?p>@tf=P
zeDDo-D_v;;n<p$@Y88dPx==mVaM<|qN9h3Oi%;6PaTK^N_W2OOlxR-I105z?E{icb
zapFYEboM|q^U>F`Ucu8{j>BYUXXFeGbqeX!n*X7kfOpQEWUHp_j*bDwBpOO9E7OV1
zzma$sv8{v>R;(;lPP3OU?<Lqc{OaviExBH?TR+I+;9+r!-@Z*j0o}Pvm%G{7zd1NG
z<R@AuJg-o`^j|mCi<feVd!W-+hc`OBf3l0txW5#Yl<1fy@w;o(Bv2>6a70V=hU~q#
zpBma?6qlBo)g)4CZ&UYfAu;El#Ju}FJ=f5jznK_151S+}|A6%K`sbGwB(P^NQvuYR
z*n;~2vvU|?7XY;;{Bi##%rX62I>LC89GPouJW$*T=wR5eVMXQTTJRYnXytH)ZXf7R
zE~=JRAW`HEC|y{zbLVup!Xla_4Q9>pSVhqdM2O;_$?3j;{G!l^>(~qNUxWCo<l{%<
zk9n->oA0vy`#-{M_Jyb9^o)rNxp89xX7()$Eay<|?|<pi=!}dEzE5`GK;!b{a6iA^
zjKh(|JN`alf*-Xe$LZA=RQ?2w5uLrx*<f-=;(PE|^;-<x;K6*49-RYfpm&10rSjuh
zb`5qep}Ku7kCZP|?8g<HO&4TOui=igY<kE_SmzhwxoPiBF`f3qVna>}9e!$RYfBIT
zm62LgR95yPuR!KL025B$a4i}vxzs@uHue&D0U1gey!~m!A%c2fR0@F_UinvOqjebk
z;AzhOM8E-p=cK%g&3O+_O#j;2&X_?y={F0rK~uKwW8(+H^MIaq!{5N(360E~H*byv
z2d7iB$cb1(@$B1CvufyhTKw+aea_E|>l=Q<$0`2)yF0uESG|F<xA1q7iXbND!5n4k
z{T}EN2G|R~2Js4+8idVZZ|zr0Rpj(EgkeDjj+|;)Y>(}{&Fb0O+mDHMj*0jmsJw+=
z?7hee^cXpHW)*K!&$-sokuD@|cV%VJ0d(s({y}BJ`K<L2!&39+rJ8y!bvtn2K)gZr
zrL_1&jqvmdnqNYHIF}cwSD3EQPGGZ}H*cykNDm)jZr{&t!)W7MdCQ+WP)d=Un(7-I
ztc_Vfq&hZ7h9`CM!>LXS#J1<y;SNivT|OHY7R4+Y42|2}<m*?i{E?fRo1C0%^QBxD
zC%Se@^J5d$YBr$301XI7m>^p^2MdM^lQDVw=aKZx69kq-#o+orm}~zHRWQd>Y3^ne
zx7Cw<^;_S*4{_e}wX{q*4PP4@O;zRP)z<Dy8D?!^F&r?<fZr#C3kYQ<*_L#G7cQZ`
zYkW#QZn^yPTF{S@A8j4m5!xg+mMcSO#iEkek7t?2j`b!}_KO>>UO!)U2wil1d}lXo
zIki2TbEBa}Nh99`9kAiXTUI;`?J7Y{C$&V+HAXXQ^J}r2Ft8*{EbOmvMBu6+Y;BpH
z78On8nte{n{f^B_OZVC<pB9bU@~e)^4<5w)9XkgFruLvFY{3*R@9Vd3#qZzmW98^Y
zww65y)+s$yD(WpIx6o}y7{$s$KMAS`{m@*nK8`XJoGHN_ADfgx>Mk$d;SbbM_fk^!
za6PgE2Bg!au$>@<2nX2>g4z^$K0>*4ri#yktCK*5$T<`bA3ofUCaygElnDJO999bS
zDJCP5vK%&GyW6A?>1ez){N$Oe^=lyjNm~&%2=9+m#1p9_o8HeQT;mPPo;tB@jN<Jd
zZ#n?_2S2+fBuwKduxF3(C`&(n%pjIoi1a_)z5rjX)%zIE;Z<Q}Wu>?UkPbKNBm5im
zI1Qcdf<ZvF?WqLng!?T;HZV4Sf}t@(H%!P#q6{o5F5bn;QS_HeM20AurF_6wdAalx
z1D)YIkK11;I6nB&vM?E{g3hTgkUqo?mv<v7P$S)ekl;_+3)tA&;^{^9Ao3^w3Nx<k
z;u5-4AG5Wf4ZJJ$p$jx6VtF?#`-rsci(Negb;}9j2`;5O6IN{jDe4qkWZS~HG8AEj
zXZ-3*mo7P>s&c!#)^0vuu!ru$krfort0tU8)0VP5T1=_b*Y+2AjOOC*u2<y?!5%c5
zYPaO3SLk>F_^6(WjE)9VOpq&lVyLfQ1Im3uZnB$Hdy#w8cg~vsNHC~5LGij^S<vSQ
zb@klq?z4rxUqlcAj40w%;GTQoB<epEE(YII*GCDbJBROe&6BIweln+m=X&~d-+=>y
zwkEw+xZL8A4^=~_s@<+ZXavz5Ftx6xCi3gR>cg(hzhf3ZSHAE??p8x2Q-YsnbJO0U
zyZ8i<J(&1NV4%KYMZ9mCk^Az+EAGdK?-sz!&WolB+}uoU?QSA3f;F|MLiVIub#yet
z>)d#%*h>+vKu<l0cwJC1kWht9)2VZ3C3{!o#!BC|!((T!v5a9Hglg5-2ag}8@KYmS
zWf?9{s)9hjj(1I-M23Bp7?nnrk~Gr(4@_@Jei>KYnp0X*q8&f4z_rz-z^P0@xL|GV
zbgPg%UVkJ$pcYf0RnXhSRfQ0TuqQLu)J~Z?Rghh;>{WzW2^#C6)$k#>++@D9Lc}P$
zB&DWAYyRF__Su!yA!s2X2fkLUZq+g{=tnE4j<f<H-VkYo;<7RiGP5a7UAuJAT)6Np
zl9Ax4PCn#*e2yHsjcb0GmKImq%$g>cg~ZR_8NEKMB)V#@XFe5U1Zdi!J+-c}O;4$Q
zXcB`9qph<xdx>n*Dp$l1P9>fr_IY`kp|jQwRNJ%eq|;t3Z8iZZxuAs)t++z?H<zDy
zb8NqZy<8;c*RR|-f8dVlSwi;0L!L+3yr5g)D}}~BqL4sUgaCZMioK>GAr}u}2<B51
z5LKF{y<apTi(RID1)B9!K+)(6yDU`YH5V`bL$_Pfl?|KB%w})eq=NgfAlr>siK4X)
z#m%PWrDbKQ7?_3TW7wzwgZd@E3g%kLm#oSfGc)(7s4d~2D12Ot+Fa7iL??C2c-vAb
z7<$qD<kc-FUuHbkhzOm8>rc=L387;9+}eCpxVp8sVMY66@7xZY{@Vskw36-4u+^w<
z#;6|x@o?$ps8rTkSkxkJyhdptH|}xp;IE{?&qC#KG|n7QmJ_SxUB2jt1@MKMYPaWM
zdcJ~Vz4KgYx0Xbs+d}(-&gLLyioDehrcV&zJ;9i6T)Cj0v{;N)AxXX^H~*5B_Ru)Q
z4El8vCJZ|^+-Y$I15?@&2$6M-IzmdU<;R+Kkce(zk3LUpW23ywe`M{Hd<`9*W*nbz
zae0HlWxacMyiE&~^p|uH?lO}Q0i>;Ddw>n(KlbVV28Mb^C988<MB+}nbB=}+`=Z^x
zg9qQ@78L8n5)z}P78IOBk`xt<w&W>|2Y5DSHNq;Z5i(~#r`7&Q-Ax=+V(oE12pjXK
z)J$N)Its6#f7PTN!gIxX+CGUe0oWxG+b632^t}}3-N(qL>Lu-*1}<^v-?%Dhx=UQj
z>uAUoL`AlOR;0>q^@SQ(nSIBO^;r^7l=1lSjY;(+MA`hOG31bTK{xXA@~+~D3n==K
zU7$(4mHjymMM~)YAL);cQf&;;*>o6<zcCmRLZ#cUU!MT{MY6}lv==%FJ;tF3&`p9q
zVpY7H8&p9DFn3g>=z4kTs`ednP?H@#4j`gqH(aA&Ll6cOM4w-IW#AG$DNgP3l8vom
zy>KaGX=zVyl)FI4Z&<ueojREigU?)J7lTU<?U(uT<#7(~8Y`>El%*fw-iXrCaPB4e
zxyG$vW=It6Xg3qs66u~#g0;hK;LW#tgTBuMGXsf-0J6zebI}$rCro1wv#FxmkK<tx
zr6#A~E{vy+6d|7weC?n$kVB;*-iHQ4Y1q)w(Mh#w$|{D@>83$AUqISj?6^ppZ*?Km
z?xry1*|Sv45YYNJ=0$7*Aa+}3S_cZg&Z6su@Xb7vPQCn(l75pOJbd_GEQpjFjbbQf
zSrB2?L=)C(?%UML>bvXvzs_WB9St)DV;OYgQ(mO#_n1AVtS~A%`W?#n@s}DJxWi=7
z_xi|74<HgC2De3ah!kG`MZoN^i&OsAF??Fkl5t|r*%euBX)C2Z&d$y>1-wffsegTa
z&&=sfD!>c~uYN?KO!ATk_5&IRqdrW2Gv+qp`@gU*l)sY`?9sn}e}Nl_S(EA2CO>|C
z4=lHKfa)&Ry|}W{NHJ83w0|ZKPjD*$4w^|lL4p4LbJz@#Y-be>RALSL5krD5u98?3
zet{rQBqJ$krum!4vuDp_ed8rGT`q=zGb4f^{b#mr9TAWMF_{T}QCXN2LTo-Nwv8><
z>AFE$FkTO=Ud6euv<~8Feq1avRyl$8{}vL6ELOCxf*U_4UYgs^t%@9~bK=K4fBxOC
zKj8bn|0MgXIG!<CFmlYGy>vlFID7)<p~`0txwn}vVNJ85yM7RK_OutzVo57{Yn}V?
zKJi|{_oF`(ZU3+T^vlc@E9Bt!{^102NDCrcfinl#;N)y4u0p5PzP`S`8D0H3OE|W?
z{Cr<OKixHv2FK^U*?G1t=4o3rNzwcFn&7<$j~<os0k_Or(I<$2g2MF=2Jn#(Ah1H&
z696E0OwK>i8~ylqyBRvWSGKJyzi4B&y7wAxZC%DrfVk+#m=1FJmT?7)BEkx-wXZTh
hAK^6TLVT;V!mU|*R2C)<7jmrma~I9I{HMw8{{UN$W_$nu

diff --git a/articles/index.html b/articles/index.html
index d2a2e2f..39f73ce 100644
--- a/articles/index.html
+++ b/articles/index.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/authors.html b/authors.html
index 32a4c4f..dbf9d1e 100644
--- a/authors.html
+++ b/authors.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
@@ -58,13 +58,13 @@ <h2 id="citation">Citation</h2>
 
       <p>Borman T, Sannikov A, Lahti L (2024).
 <em>HoloFoodR: R interface to EBI HoloFood resource</em>.
-R package version 0.99.10, <a href="https://github.com/EBI-Metagenomics/HoloFoodR" class="external-link">https://github.com/EBI-Metagenomics/HoloFoodR</a>.
+R package version 0.99.11, <a href="https://github.com/EBI-Metagenomics/HoloFoodR" class="external-link">https://github.com/EBI-Metagenomics/HoloFoodR</a>.
 </p>
       <pre>@Manual{,
   title = {HoloFoodR: R interface to EBI HoloFood resource},
   author = {Tuomas Borman and Artur Sannikov and Leo Lahti},
   year = {2024},
-  note = {R package version 0.99.10},
+  note = {R package version 0.99.11},
   url = {https://github.com/EBI-Metagenomics/HoloFoodR},
 }</pre>
     </div>
diff --git a/index.html b/index.html
index 90480dc..57b70c6 100644
--- a/index.html
+++ b/index.html
@@ -22,7 +22,7 @@
 
     <a class="navbar-brand me-2" href="index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/pkgdown.yml b/pkgdown.yml
index dd87b43..e8d8ee2 100644
--- a/pkgdown.yml
+++ b/pkgdown.yml
@@ -4,4 +4,4 @@ pkgdown_sha: ~
 articles:
   case_study: case_study.html
   HoloFoodR: HoloFoodR.html
-last_built: 2024-10-24T12:51Z
+last_built: 2024-10-24T13:27Z
diff --git a/reference/HoloFoodR.html b/reference/HoloFoodR.html
index b635660..1f9eab7 100644
--- a/reference/HoloFoodR.html
+++ b/reference/HoloFoodR.html
@@ -17,7 +17,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/addMGnify.html b/reference/addMGnify.html
index f3cebc0..0a3274b 100644
--- a/reference/addMGnify.html
+++ b/reference/addMGnify.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/doQuery.html b/reference/doQuery.html
index 97b2b22..e2d6312 100644
--- a/reference/doQuery.html
+++ b/reference/doQuery.html
@@ -9,7 +9,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getData.html b/reference/getData.html
index 97f6158..17c4b0e 100644
--- a/reference/getData.html
+++ b/reference/getData.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getMetaboLights.html b/reference/getMetaboLights.html
index 4de0528..53bc7cd 100644
--- a/reference/getMetaboLights.html
+++ b/reference/getMetaboLights.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/getResult.html b/reference/getResult.html
index 9ef5248..15805a9 100644
--- a/reference/getResult.html
+++ b/reference/getResult.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/reference/index.html b/reference/index.html
index 8fd53fd..965ff96 100644
--- a/reference/index.html
+++ b/reference/index.html
@@ -7,7 +7,7 @@
 
     <a class="navbar-brand me-2" href="../index.html">HoloFoodR</a>
 
-    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.10</small>
+    <small class="nav-text text-muted me-auto" data-bs-toggle="tooltip" data-bs-placement="bottom" title="">0.99.11</small>
 
 
     <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbar" aria-controls="navbar" aria-expanded="false" aria-label="Toggle navigation">
diff --git a/search.json b/search.json
index 645f3de..17ff57c 100644
--- a/search.json
+++ b/search.json
@@ -1 +1 @@
-[{"path":"/articles/HoloFoodR.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFoodR package designed ease access EBI’s HoloFoodR resource, allowing searching retrieval multiple datasets downstream analysis. HoloFood database encompass metagenomics data; however, data stored within MGnify database. packages offer analogous functionalities, streamlining integration data enhancing accessibility. TreeSummarizedExperiment","code":""},{"path":"/articles/HoloFoodR.html","id":"installation","dir":"Articles","previous_headings":"","what":"Installation","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFoodR hosted Bioconductor, can installed using via BiocManager.","code":"BiocManager::install(\"HoloFoodR\")"},{"path":"/articles/HoloFoodR.html","id":"load-the-package","dir":"Articles","previous_headings":"","what":"Load the package","title":"HoloFoodR: interface to HoloFoodR database","text":"installed, HoloFoodR made available usual way.","code":"library(HoloFoodR) #> Loading required package: MultiAssayExperiment #> Loading required package: SummarizedExperiment #> Loading required package: MatrixGenerics #> Loading required package: matrixStats #>  #> Attaching package: 'MatrixGenerics' #> The following objects are masked from 'package:matrixStats': #>  #>     colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse, #>     colCounts, colCummaxs, colCummins, colCumprods, colCumsums, #>     colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs, #>     colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats, #>     colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds, #>     colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads, #>     colWeightedMeans, colWeightedMedians, colWeightedSds, #>     colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet, #>     rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods, #>     rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps, #>     rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins, #>     rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks, #>     rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars, #>     rowWeightedMads, rowWeightedMeans, rowWeightedMedians, #>     rowWeightedSds, rowWeightedVars #> Loading required package: GenomicRanges #> Loading required package: stats4 #> Loading required package: BiocGenerics #>  #> Attaching package: 'BiocGenerics' #> The following objects are masked from 'package:stats': #>  #>     IQR, mad, sd, var, xtabs #> The following objects are masked from 'package:base': #>  #>     anyDuplicated, aperm, append, as.data.frame, basename, cbind, #>     colnames, dirname, do.call, duplicated, eval, evalq, Filter, Find, #>     get, grep, grepl, intersect, is.unsorted, lapply, Map, mapply, #>     match, mget, order, paste, pmax, pmax.int, pmin, pmin.int, #>     Position, rank, rbind, Reduce, rownames, sapply, saveRDS, setdiff, #>     table, tapply, union, unique, unsplit, which.max, which.min #> Loading required package: S4Vectors #>  #> Attaching package: 'S4Vectors' #> The following object is masked from 'package:utils': #>  #>     findMatches #> The following objects are masked from 'package:base': #>  #>     expand.grid, I, unname #> Loading required package: IRanges #> Loading required package: GenomeInfoDb #> Loading required package: Biobase #> Welcome to Bioconductor #>  #>     Vignettes contain introductory material; view with #>     'browseVignettes()'. To cite Bioconductor, see #>     'citation(\"Biobase\")', and for packages 'citation(\"pkgname\")'. #>  #> Attaching package: 'Biobase' #> The following object is masked from 'package:MatrixGenerics': #>  #>     rowMedians #> The following objects are masked from 'package:matrixStats': #>  #>     anyMissing, rowMedians #> Loading required package: TreeSummarizedExperiment #> Loading required package: SingleCellExperiment #> Loading required package: Biostrings #> Loading required package: XVector #>  #> Attaching package: 'Biostrings' #> The following object is masked from 'package:base': #>  #>     strsplit"},{"path":"/articles/HoloFoodR.html","id":"functionalities","dir":"Articles","previous_headings":"","what":"Functionalities","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFoodR offers three functions doQuery(), getResult() getData() can utilized search fetch data HoloFood database. tutorial, demonstrate search animals, subset animals based whether specific sample type, finally fetch data samples. Note can done doQuery() getResult() (getData() getResult()) utilizing query filters. tutorial demonstrating functionality package. Additionally, package includes getMetaboLights() function can utilized retrieve metabolomic data MetaboLights database.","code":""},{"path":"/articles/HoloFoodR.html","id":"search-data","dir":"Articles","previous_headings":"Functionalities","what":"Search data","title":"HoloFoodR: interface to HoloFoodR database","text":"search animals, genome catalogues, samples viral catalogues, can use doQuery() function. can also use getData() doQuery() optimized searching datatypes. example, instead nested list sample types doQuery() returns sample types presence/absence table convenient. search animals, subset based whether include histological samples. Note can done also using query filters. doQuery() returns data.frame including information type data searched.","code":"animals <- doQuery(\"animals\", max.hits = 100) animals <- animals[ animals[[\"histological\"]], ]  colnames(animals) |> head() #> [1] \"accession\"            \"system\"               \"canonical_url\"        #> [4] \"histological\"         \"host_genomic\"         \"inflammatory_markers\""},{"path":"/articles/HoloFoodR.html","id":"get-data","dir":"Articles","previous_headings":"Functionalities","what":"Get data","title":"HoloFoodR: interface to HoloFoodR database","text":"Now information animal histological samples. Let’s get data animals know sample IDs fetch. returned value getData() function list. can data also data.frame specify flatten = TRUE. data information animals including samples drawn . elements list data.frames. example, “samples” table contains information samples drawn animals specified input. Now can collect sample IDs.","code":"animal_data <- getData(     accession.type = \"animals\", accession = animals[[\"accession\"]]) samples <- animal_data[[\"samples\"]]  colnames(samples) |> head() #> [1] \"accession\"        \"title\"            \"sample_type\"      \"animal\"           #> [5] \"canonical_url\"    \"metagenomics_url\" sample_ids <- unique(samples[[\"accession\"]])"},{"path":"/articles/HoloFoodR.html","id":"get-data-on-samples","dir":"Articles","previous_headings":"Functionalities","what":"Get data on samples","title":"HoloFoodR: interface to HoloFoodR database","text":"get data samples, can utilize getResult() function. returns data MultiAssayExperiment (MAE) format. MAE object stores individual omics TreeSummarizedExperiment (TreeSE) objects. TreeSE, column represents sample rows represent features.","code":"mae <- getResult(sample_ids) #> Warning: Data for the following samples cannot be found. The sample types are metagenomic_assembly, host_genomic, transcriptomic and metatranscriptomic. (Note that metagenomic assemblies can be found from the MGnify database. See MGnifyR package.): #> 'SAMEA10130025', 'SAMEA13389405', 'SAMEA13389406', 'SAMEA13901590', 'SAMEA13901591', 'SAMEA13929779', 'SAMEA7697591', 'SAMEA10130091', 'SAMEA13389692', 'SAMEA13389693', 'SAMEA13901708', 'SAMEA7571845', 'SAMEA10158030', 'SAMEA13389419', 'SAMEA13389420', 'SAMEA13901594', 'SAMEA13901595', 'SAMEA13929781', 'SAMEA7697592', 'SAMEA10130039', 'SAMEA13389496', 'SAMEA13389497', 'SAMEA13901618', 'SAMEA13901619', 'SAMEA13929785', 'SAMEA7571815', 'SAMEA10130112', 'SAMEA13389794', 'SAMEA13389795', 'SAMEA13901758', 'SAMEA13901759', 'SAMEA13929811', 'SAMEA7571864', 'SAMEA10158022', 'SAMEA13389146', 'SAMEA13389147', 'SAMEA13901511', 'SAMEA13901512', 'SAMEA13929767', 'SAMEA7571777', 'SAMEA10130019', 'SAMEA13389353', 'SAMEA13389354', 'SAMEA13389355', 'SAMEA13901574', 'SAMEA13901575', 'SAMEA14095991', 'SAMEA7722475', 'SAMEA10130101', 'SAMEA13389738', 'SAMEA13389739', 'SAMEA13901730', 'SAMEA13901731', 'SAMEA13929802', 'SAMEA7571856', 'SAMEA10455480', 'SAMEA13389220', 'SAMEA10129993', 'SAMEA13389183', 'SAMEA13389184', 'SAMEA13901520', 'SAMEA13901521', 'SAMEA7697579', 'SAMEA10130017', 'SAMEA13389345', 'SAMEA13389346', 'SAMEA13901571', 'SAMEA13901572', 'SAMEA13929772', 'SAMEA7571801', 'SAMEA10130113', 'SAMEA13389807', 'SAMEA13389808', 'SAMEA13901762', 'SAMEA13901763', 'SAMEA13929813', 'SAMEA7571866', 'SAMEA10455481', 'SAMEA13389227', 'SAMEA10455479', 'SAMEA13389169', 'SAMEA10130020', 'SAMEA13389357', 'SAMEA13389358', 'SAMEA13901576', 'SAMEA13901577', 'SAMEA13929773', 'SAMEA7697587', 'SAMEA10130100', 'SAMEA13389734', 'SAMEA13389735', 'SAMEA13901728', 'SAMEA13901729', 'SAMEA13929801', 'SAMEA7697622', 'SAMEA10130016', 'SAMEA13389342', 'SAMEA13389343', 'SAMEA13901569', 'SAMEA13901570', 'SAMEA13929771', 'SAMEA7571800', 'SAMEA10130040', 'SAMEA13389503', 'SAMEA13389504', 'SAMEA13901620', 'SAMEA13901621', 'SAMEA13929786', 'SAMEA7571816', 'SAMEA10129979', 'SAMEA13389081', 'SAMEA13389082', 'SAMEA13901489', 'SAMEA13901490', 'SAMEA7571769', 'SAMEA10130002', 'SAMEA13389243', 'SAMEA13389244', 'SAMEA13901540', 'SAMEA13901541', 'SAMEA7697582', 'SAMEA10129985', 'SAMEA13389133', 'SAMEA13389134', 'SAMEA13901505', 'SAMEA13901506', 'SAMEA13929764', 'SAMEA7571775', 'SAMEA10455476', 'SAMEA13389110', 'SAMEA10130031', 'SAMEA13389443', 'SAMEA13389444', 'SAMEA13901602', 'SAMEA13901603', 'SAMEA7571811', 'SAMEA10130023', 'SAMEA13389395', 'SAMEA13389396', 'SAMEA13901586', 'SAMEA13901587', 'SAMEA13929777', 'SAMEA7571806', 'SAMEA10130090', 'SAMEA13389687', 'SAMEA13389688', 'SAMEA13901707', 'SAMEA7571844', 'SAMEA10130119', 'SAMEA13389832', 'SAMEA13389833', 'SAMEA13901773', 'SAMEA13929818', 'SAMEA7697633', 'SAMEA10129996', 'SAMEA13389204', 'SAMEA13389205', 'SAMEA13901529', 'SAMEA13901530', 'SAMEA7697580', 'SAMEA10130088', 'SAMEA13389677', 'SAMEA13389678', 'SAMEA13901704', 'SAMEA13929799', 'SAMEA7571843' mae #> A MultiAssayExperiment object of 8 listed #>  experiments with user-defined names and respective classes. #>  Containing an ExperimentList class object of length 8: #>  [1] BIOGENIC AMINES: TreeSummarizedExperiment with 7 rows and 35 columns #>  [2] FATTY ACIDS: TreeSummarizedExperiment with 19 rows and 57 columns #>  [3] HISTOLOGY: TreeSummarizedExperiment with 20 rows and 57 columns #>  [4] INFLAMMATORY MARKERS: TreeSummarizedExperiment with 14 rows and 58 columns #>  [5] metagenomic_assembly: TreeSummarizedExperiment with 0 rows and 53 columns #>  [6] host_genomic: TreeSummarizedExperiment with 0 rows and 53 columns #>  [7] transcriptomic: TreeSummarizedExperiment with 0 rows and 44 columns #>  [8] metatranscriptomic: TreeSummarizedExperiment with 0 rows and 16 columns #> Functionality: #>  experiments() - obtain the ExperimentList instance #>  colData() - the primary/phenotype DataFrame #>  sampleMap() - the sample coordination DataFrame #>  `$`, `[`, `[[` - extract colData columns, subset, or experiment #>  *Format() - convert into a long or wide DataFrame #>  assays() - convert ExperimentList to a SimpleList of matrices #>  exportClass() - save data to flat files mae[[1]] #> class: TreeSummarizedExperiment  #> dim: 7 35  #> metadata(0): #> assays(1): counts #> rownames(7): Cadaverin Gesamtamine (Total biogenic amines) ... Spermin #>   Tyramin #> rowData names(4): marker.name marker.type marker.canonical_url units #> colnames(35): SAMEA112906114 SAMEA112906592 ... SAMEA112906002 #>   SAMEA112906785 #> colData names(13): accession sample_type ... Project Sample code #> reducedDimNames(0): #> mainExpName: NULL #> altExpNames(0): #> rowLinks: NULL #> rowTree: NULL #> colLinks: NULL #> colTree: NULL"},{"path":"/articles/HoloFoodR.html","id":"incorporate-with-mgnify-data","dir":"Articles","previous_headings":"Functionalities","what":"Incorporate with MGnify data","title":"HoloFoodR: interface to HoloFoodR database","text":"MGnifyR package can utilized fetch metagenomics data MGnify database. MGnifyR package, can use MGnifyR::searchAnalysis() function search analyses based sample IDs . can fetch data based accession IDs. MGnifyR::getResult() returns MAE object just like HoloFoodR. However, metagenomic data points individual analyses instead samples. can harmonize data replacing analysis IDs sample IDs, can combine data single MAE. Now, MAE object linking samples various omics together, compatibility ensured single omics datasets (Tree)SummarizedExperiment format. compatibility allows us harness cutting-edge downstream analytics tools like miaverse framework support data containers seamlessly.","code":"library(MGnifyR)  mg <- MgnifyClient(useCache = TRUE)  # Get those samples that are metagenomic samples metagenomic_samples <- samples[     samples[[\"sample_type\"]] == \"metagenomic_assembly\", ]  # Get analysis IDs based on sample IDs analysis_ids <- searchAnalysis(     mg, type = \"samples\", metagenomic_samples[[\"accession\"]]) head(analysis_ids) #>  SAMEA10130025   SAMEA7697591  SAMEA10130091   SAMEA7571845  SAMEA10158030  #> \"MGYA00606535\" \"MGYA00616692\" \"MGYA00606528\" \"MGYA00616689\" \"MGYA00606518\"  #>   SAMEA7697592  #> \"MGYA00615947\" mae_metagenomic <- MGnifyR::getResult(mg, analysis_ids) mae_metagenomic #> A MultiAssayExperiment object of 6 listed #>  experiments with user-defined names and respective classes. #>  Containing an ExperimentList class object of length 6: #>  [1] microbiota: TreeSummarizedExperiment with 675 rows and 52 columns #>  [2] go-slim: TreeSummarizedExperiment with 116 rows and 52 columns #>  [3] go-terms: TreeSummarizedExperiment with 3264 rows and 52 columns #>  [4] interpro-identifiers: TreeSummarizedExperiment with 19681 rows and 52 columns #>  [5] taxonomy: TreeSummarizedExperiment with 1438 rows and 52 columns #>  [6] taxonomy-lsu: TreeSummarizedExperiment with 1856 rows and 52 columns #> Functionality: #>  experiments() - obtain the ExperimentList instance #>  colData() - the primary/phenotype DataFrame #>  sampleMap() - the sample coordination DataFrame #>  `$`, `[`, `[[` - extract colData columns, subset, or experiment #>  *Format() - convert into a long or wide DataFrame #>  assays() - convert ExperimentList to a SimpleList of matrices #>  exportClass() - save data to flat files # Get experiments from metagenomic data exps <- experiments(mae_metagenomic) # Convert analysis names to sample names exps <- lapply(exps, function(x){     # Get corresponding sample ID     sample_id <- names(analysis_ids)[ match(colnames(x), analysis_ids) ]     # Replace analysis ID with sample ID     colnames(x) <- sample_id     return(x) })  # Add to original MultiAssayExperiment mae <- c(experiments(mae), exps) mae"},{"path":"/articles/HoloFoodR.html","id":"extra-get-data-from-metabolights-database","dir":"Articles","previous_headings":"Functionalities","what":"Extra: Get data from MetaboLights database","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFood database exclusively contains targeted metabolomic data. However, provides URL addresses linking MetaboLights database, untargeted metabolomics data can accessed. retrieve data, can utilize getMetaboLights() function retrieve information available data. Moreover, returns processed metabolomic data (processed data, can also use getReturn(x, get.metabolomic=TRUE)). , retrieve processed (mapped) metabolomic data associated HoloFood. result list includes three data.frames: spectra data, can either fetch files using getMetaboLightsFile(), follow vignette guidance loading data directly object, tailored metabolomics spectra data.","code":"# Get untargeted metabolomic samples samples <- doQuery(\"samples\", sample_type = \"metabolomic\") # Get the data metabolomic <- getMetaboLights(samples[[\"metabolomics_url\"]])  # Show names of data.frames names(metabolomic) - study metadata - assay metadata - assay that includes abundance table and feature metadata"},{"path":"/articles/HoloFoodR.html","id":"session-info","dir":"Articles","previous_headings":"","what":"Session info","title":"HoloFoodR: interface to HoloFoodR database","text":"","code":"sessionInfo() #> R version 4.4.1 (2024-06-14) #> Platform: x86_64-pc-linux-gnu #> Running under: Ubuntu 22.04.5 LTS #>  #> Matrix products: default #> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  #> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 #>  #> locale: #>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               #>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     #>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    #>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  #>  [9] LC_ADDRESS=C               LC_TELEPHONE=C             #> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        #>  #> time zone: UTC #> tzcode source: system (glibc) #>  #> attached base packages: #> [1] stats4    stats     graphics  grDevices utils     datasets  methods   #> [8] base      #>  #> other attached packages: #>  [1] HoloFoodR_0.99.10               TreeSummarizedExperiment_2.13.0 #>  [3] Biostrings_2.73.2               XVector_0.45.0                  #>  [5] SingleCellExperiment_1.27.2     MultiAssayExperiment_1.31.5     #>  [7] SummarizedExperiment_1.35.4     Biobase_2.65.1                  #>  [9] GenomicRanges_1.57.2            GenomeInfoDb_1.41.2             #> [11] IRanges_2.39.2                  S4Vectors_0.43.2                #> [13] BiocGenerics_0.51.3             MatrixGenerics_1.17.0           #> [15] matrixStats_1.4.1               knitr_1.48                      #> [17] BiocStyle_2.33.1                #>  #> loaded via a namespace (and not attached): #>  [1] httr2_1.0.5             xfun_0.48               bslib_0.8.0             #>  [4] htmlwidgets_1.6.4       lattice_0.22-6          yulab.utils_0.1.7       #>  [7] vctrs_0.6.5             tools_4.4.1             generics_0.1.3          #> [10] curl_5.2.3              parallel_4.4.1          tibble_3.2.1            #> [13] fansi_1.0.6             BiocBaseUtils_1.7.3     pkgconfig_2.0.3         #> [16] Matrix_1.7-0            desc_1.4.3              lifecycle_1.0.4         #> [19] GenomeInfoDbData_1.2.13 compiler_4.4.1          treeio_1.29.1           #> [22] textshaping_0.4.0       codetools_0.2-20        htmltools_0.5.8.1       #> [25] sass_0.4.9              lazyeval_0.2.2          yaml_2.3.10             #> [28] tidyr_1.3.1             pkgdown_2.1.1           pillar_1.9.0            #> [31] crayon_1.5.3            jquerylib_0.1.4         BiocParallel_1.39.0     #> [34] DelayedArray_0.31.14    cachem_1.1.0            abind_1.4-8             #> [37] nlme_3.1-166            tidyselect_1.2.1        digest_0.6.37           #> [40] purrr_1.0.2             dplyr_1.1.4             bookdown_0.41           #> [43] fastmap_1.2.0           grid_4.4.1              cli_3.6.3               #> [46] SparseArray_1.5.45      magrittr_2.0.3          S4Arrays_1.5.11         #> [49] utf8_1.2.4              ape_5.8                 rappdirs_0.3.3          #> [52] UCSC.utils_1.1.0        rmarkdown_2.28          httr_1.4.7              #> [55] ragg_1.3.3              evaluate_1.0.1          rlang_1.1.4             #> [58] Rcpp_1.0.13             tidytree_0.4.6          glue_1.8.0              #> [61] BiocManager_1.30.25     jsonlite_1.8.9          R6_2.5.1                #> [64] systemfonts_1.1.0       fs_1.6.4                zlibbioc_1.51.2"},{"path":"/articles/case_study.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"HoloFoodR R package — case study","text":"HoloFood database large collection holo-omic multi-omic data two animal systems, chicken salmon. created HoloFood consortium. One objectives study interactions animal systems, microbiomes, feed additives optimize diet strategies farm animals. make easily accessible, creators provide Application Programming Interface (API) permits interaction programming languages, R Python. improve accessibility creating HoloFoodR package simplifies API interaction assists translating raw database data R/Bioconductor data containers, connecting vast ecosystem bioinformatics R packages. aim demonstrate HoloFoodR isolation rest ecosystem, showcase possibility data integration databases, MGnify, holds metagenomic data. Furthermore, provide workflow data exploration advanced machine learning multi-omics, offering practical example readers. main study questions :","code":"- Do treatment associate with salmons' gut flora? - Do larger salmons have different gut microbiota compared to smaller ones? - Do gut microbiota have association with muscle tissue fatty acids? # List of packages that we need packages <- c(     \"dplyr\", \"DT\", \"ggsignif\", \"HoloFoodR\", \"MGnifyR\", \"mia\", \"miaViz\", \"MOFA2\",     \"patchwork\", \"reticulate\", \"scater\" )  # Load all packages into session. Stop if there are packages that were not # successfully loaded pkgs_not_loaded <- !sapply(packages, require, character.only = TRUE) #> Loading required package: dplyr #>  #> Attaching package: 'dplyr' #> The following objects are masked from 'package:stats': #>  #>     filter, lag #> The following objects are masked from 'package:base': #>  #>     intersect, setdiff, setequal, union #> Loading required package: DT #> Loading required package: ggsignif #> Loading required package: HoloFoodR #> Loading required package: MultiAssayExperiment #> Loading required package: SummarizedExperiment #> Loading required package: MatrixGenerics #> Loading required package: matrixStats #>  #> Attaching package: 'matrixStats' #> The following object is masked from 'package:dplyr': #>  #>     count #>  #> Attaching package: 'MatrixGenerics' #> The following objects are masked from 'package:matrixStats': #>  #>     colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse, #>     colCounts, colCummaxs, colCummins, colCumprods, colCumsums, #>     colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs, #>     colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats, #>     colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds, #>     colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads, #>     colWeightedMeans, colWeightedMedians, colWeightedSds, #>     colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet, #>     rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods, #>     rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps, #>     rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins, #>     rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks, #>     rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars, #>     rowWeightedMads, rowWeightedMeans, rowWeightedMedians, #>     rowWeightedSds, rowWeightedVars #> Loading required package: GenomicRanges #> Loading required package: stats4 #> Loading required package: BiocGenerics #>  #> Attaching package: 'BiocGenerics' #> The following objects are masked from 'package:dplyr': #>  #>     combine, intersect, setdiff, union #> The following objects are masked from 'package:stats': #>  #>     IQR, mad, sd, var, xtabs #> The following objects are masked from 'package:base': #>  #>     anyDuplicated, aperm, append, as.data.frame, basename, cbind, #>     colnames, dirname, do.call, duplicated, eval, evalq, Filter, Find, #>     get, grep, grepl, intersect, is.unsorted, lapply, Map, mapply, #>     match, mget, order, paste, pmax, pmax.int, pmin, pmin.int, #>     Position, rank, rbind, Reduce, rownames, sapply, saveRDS, setdiff, #>     table, tapply, union, unique, unsplit, which.max, which.min #> Loading required package: S4Vectors #>  #> Attaching package: 'S4Vectors' #> The following objects are masked from 'package:dplyr': #>  #>     first, rename #> The following object is masked from 'package:utils': #>  #>     findMatches #> The following objects are masked from 'package:base': #>  #>     expand.grid, I, unname #> Loading required package: IRanges #>  #> Attaching package: 'IRanges' #> The following objects are masked from 'package:dplyr': #>  #>     collapse, desc, slice #> Loading required package: GenomeInfoDb #> Loading required package: Biobase #> Welcome to Bioconductor #>  #>     Vignettes contain introductory material; view with #>     'browseVignettes()'. To cite Bioconductor, see #>     'citation(\"Biobase\")', and for packages 'citation(\"pkgname\")'. #>  #> Attaching package: 'Biobase' #> The following object is masked from 'package:MatrixGenerics': #>  #>     rowMedians #> The following objects are masked from 'package:matrixStats': #>  #>     anyMissing, rowMedians #> Loading required package: TreeSummarizedExperiment #> Loading required package: SingleCellExperiment #> Loading required package: Biostrings #> Loading required package: XVector #>  #> Attaching package: 'Biostrings' #> The following object is masked from 'package:base': #>  #>     strsplit #> Loading required package: MGnifyR #>  #> Attaching package: 'MGnifyR' #> The following objects are masked from 'package:HoloFoodR': #>  #>     doQuery, getData, getResult #> Loading required package: mia #> This is mia version 1.13.47 #> - Online documentation and vignettes: https://microbiome.github.io/mia/ #> - Online book 'Orchestrating Microbiome Analysis (OMA)': https://microbiome.github.io/OMA/docs/devel/ #>  #> Attaching package: 'mia' #> The following objects are masked from 'package:dplyr': #>  #>     full_join, inner_join, left_join, right_join #> Loading required package: miaViz #> Loading required package: ggplot2 #> Loading required package: ggraph #>  #> Attaching package: 'miaViz' #> The following object is masked from 'package:mia': #>  #>     plotNMDS #> Loading required package: MOFA2 #>  #> Attaching package: 'MOFA2' #> The following object is masked from 'package:stats': #>  #>     predict #> Loading required package: patchwork #> Loading required package: reticulate #> Loading required package: scater #> Loading required package: scuttle pkgs_not_loaded <- names(pkgs_not_loaded)[pkgs_not_loaded] if (length(pkgs_not_loaded) > 0) {     stop(\"Error in loading the following packages into the session: '\",         paste0(pkgs_not_loaded, collapse = \"', '\"), \"'\") }"},{"path":"/articles/case_study.html","id":"import-data","dir":"Articles","previous_headings":"","what":"Import data","title":"HoloFoodR R package — case study","text":"start workflow data retrieval. use salmon data associated fatty acid metagenomic amplicon example.","code":""},{"path":"/articles/case_study.html","id":"retrieve-holofood-data","dir":"Articles","previous_headings":"Import data","what":"Retrieve HoloFood data","title":"HoloFoodR R package — case study","text":"First , query HoloFood database retrieve salmon accession numbers. Next, can retrieve data associated salmon. data returned list sample accession numbers associated salmons. example, metagenomic amplicon samples, SAMEA112750580 genomic sample, SAMEA9068691. can use accession numbers fetch data associated sample type store experiments MultiAssayExperiment (MAE) object.","code":"# Get salmon samples salmons <- HoloFoodR::doQuery(\"animals\", system = \"salmon\", use.cache = TRUE)  # Get only the data that has both metagenomic amplicon and fatty acid # data salmons <- salmons |>     filter(fatty_acids == TRUE & metagenomic_amplicon == TRUE)  colnames(salmons) # Get salmon data salmon_data <- HoloFoodR::getData(     accession.type = \"animals\",     accession = salmons[[\"accession\"]],     use.cache = TRUE )  # Get salmon samples salmon_samples <- salmon_data[[\"samples\"]]  # Get sample IDs salmon_sample_ids <- unique(salmon_samples[[\"accession\"]])  head(salmon_sample_ids) # Get salmon <- experiments as MAE object mae <- HoloFoodR::getResult(     salmon_sample_ids,     use.cache = TRUE )"},{"path":"/articles/case_study.html","id":"fetch-metagenomic-data-from-mgnify","dir":"Articles","previous_headings":"Import data","what":"Fetch metagenomic data from MGnify","title":"HoloFoodR R package — case study","text":"HoloFood database include data metagenomic data. data can retrieved MGnify portal. purpose, use MGnifyR package, similar fashion HoloFoodR, allows simple interaction MGnify API. salmon_analysis_ids character vector holds associations HoloFood metagenomic amplicon accession numbers (SAMEAxxxxxx) counterparts MGnify database (MGYAxxxxxx). Data fetched MGnify MGnify-specific identifiers. first rename samples HoloFood specific ID add data MultiAssayExperiment combining data. Now retrieved data interested workflow.","code":"# Create MGnify object mg <- MgnifyClient(     useCache = TRUE,     cacheDir = \".MGnifyR_cache\" )  # Select only metagenomic_amplicon sample type metagenomic_salmon_samples <- salmon_samples |>     filter(sample_type == \"metagenomic_amplicon\")  # Search for sample IDs in MGnify database salmon_analysis_ids <- searchAnalysis(     mg,     type = \"samples\",     metagenomic_salmon_samples[[\"accession\"]] ) # Get metagenomic taxonomic data for salmon from MGnify tse <- MGnifyR::getResult(     mg,     accession = salmon_analysis_ids,     get.func = FALSE ) # Add MGnify results to HoloFood data mae <- addMGnify(tse, mae) #> Warning: 'experiments' dropped; see 'drops()' #> harmonizing input: #>   removing 189 sampleMap rows not in names(experiments) #> harmonizing input: #>   removing 189 sampleMap rows not in names(experiments)"},{"path":"/articles/case_study.html","id":"data-preprocess","dir":"Articles","previous_headings":"","what":"Data preprocess","title":"HoloFoodR R package — case study","text":"Data cleaning one time-consuming important steps data analysis. instance, need handle missing data, transform data assays, agglomerate data. next steps, : Filter data Ensure data correct format analysis Agglomerate data Transform data","code":""},{"path":"/articles/case_study.html","id":"wrangle-the-data","dir":"Articles","previous_headings":"Data preprocess","what":"Wrangle the data","title":"HoloFoodR R package — case study","text":"see upset plot summarizes available experiments samples overlap .  demonstration purposes, focus investigating fatty acids metagenomic data (fatty_acids_mg metagenomic_amplicon) within trial performed HoloFood consortium. trial health effects fermented seaweed added diet salmons. values fatty acids detection thresholds. assume zeroes. Moreover, data includes feature just states fatty acids collected. remove feature assay contains numeric values. Moreover, wrangle sample metadata includes necessary information. want analyze effect treatment, must analyze salmons day 60 salmons sampled experimental period. contrast, day-0 salmons sampled immediately beginning trial. select samples drawn day 60. Now data includes samples taken day 60. animal microbiota data fatty acid data.","code":"upset_plot <- upsetSamples(mae) upset_plot # Harmonize experiment names names(mae) <- names(mae) |>     tolower() |>     gsub(pattern = \" \", replacement = \"_\")  # Fetch only experiments that we need mae <- mae[, , c(\"fatty_acids_mg\", \"metagenomic\")] #> Warning: 'experiments' dropped; see 'drops()' #> harmonizing input: #>   removing 934 sampleMap rows not in names(experiments) names(mae) <- c(\"fatty_acids\", \"metagenomic\") # Filter MAE object to include only Trial A mae <- mae[, colData(mae)[[\"Trial code\"]] == \"SA\", ] # From metabolomic data, remove organ-fatty acids row because it only contains a # string value \"muscle\" which denotes where the sample was drawn. tse <- mae[[1]] tse <- tse[!(rowData(tse)[[\"marker.name\"]] %in% c(\"Organ-fatty acids\")), ] mae[[1]] <- tse  # Transform matrix to numeric. Some values are \"< 0.01\" # If a number is < 0.01, assume it to be 0 assay <- assay(mae[[1]], \"counts\") assay[assay == \"<0.01\"] <- 0 assay <- apply(assay, c(1, 2), function(x) as.numeric(gsub(\",\", \".\", x)))  # Reassign assay back to MAE assay(mae[[1]], \"counts\") <- assay # Add treatment groups colData(mae)[[\"treatment_group\"]] <- ifelse(   colData(mae)[[\"Treatment concentration\"]]>0, \"treatment\", \"control\")  # Add salmon host mass to MAE colData colData(mae[[2]])[[\"host.gutted.mass\"]] <- as.numeric(     colData(mae[[2]])[[\"host.gutted.mass\"]]) colData(mae[[2]])[[\"mass_scaled\"]] <- scale(   colData(mae[[2]])[[\"host.gutted.mass\"]]) merged_data <- merge(     x = colData(mae),     y = colData(mae[[2]])[, c(\"animal\", \"host.gutted.mass\", \"mass_scaled\")],     all.x = TRUE, by = \"animal\" ) rownames(merged_data) <- merged_data[[\"animal\"]] # Order and add back colData(mae) <- merged_data[match(     colData(mae)[[\"animal\"]], merged_data[[\"animal\"]]), ]  # Add animal metadata to separate experiments mae[[1]] <- getWithColData(mae, 1) #> Warning: Ignoring redundant column names in 'colData(x)': canonical_url, #> marker.type mae[[2]] <- getWithColData(mae, 2) #> Warning: Ignoring redundant column names in 'colData(x)': canonical_url, marker.type, #>   mass_scaled # Subset by taking only samples from day 60 samples <- colnames(mae[[2]])[mae[[2]]$trial.timepoint == 60] mae[[2]] <- mae[[2]][, samples] #> harmonizing input: #>   removing 15 sampleMap rows with 'colname' not in colnames of experiments # Make sure that we have same set of samples from each experiment mae <- intersectColumns(mae)"},{"path":"/articles/case_study.html","id":"filtering-and-agglomeration","dir":"Articles","previous_headings":"Data preprocess","what":"Filtering and agglomeration","title":"HoloFoodR R package — case study","text":"Next, can agglomerate features prevalence reduce number low-abundant taxa contaminants. First, visualize prevalence distribution taxa histogram decide prevalence threshold use. use 0.1% detection level filter extremely low-abundant genera.  can also look raw prevalence numbers Mycoplasma present samples, surprising genus found one common salmon intestine (see Zarkasi et al. (2014)). use thresholds 20% 0.1% prevalence detection, respectively. Moreover, agglomerate data include 19 abundant genera. Due limited number samples, also filter fatty acid data include fatty acids show variation within dataset. rationale fatty acid vary, exhibit differences groups. can find good threshold cut-histogram standard deviations fatty acid abundances.  apply filtering threshold 0.5 exclude fatty acids exhibit sufficient variation dataset. Since standard deviations 1, reasonable number want exclude many fatty acids. detailed analysis, pick certain fatty acids well-established biological relevance. include 22:6n-3, 20:5n-3, 18:3n-3, 20:4n-6, 18:2n-6, 18:1n-9, 16:0, 18:0.","code":"# Add relative transforrmed data mae[[2]] <- transformAssay(mae[[2]], method = \"relabundance\")  prevalence <- getPrevalence(     mae[[2]],     rank = \"Genus\",     assay.type = \"relabundance\",     na.rm = TRUE,     sort = TRUE,     detection = 0.1 / 100 )  # Exclude microbes with 0 prevalence prevalence <- prevalence[prevalence != 0]  hist(prevalence, main = \"Prevalence of microbial taxa\",      xlab = \"Prevalence\") # Sort prevalence in decreasing order sort(prevalence, decreasing = TRUE) |> head(10) #>     Mycoplasma Photobacterium     Aliivibrio  Cetobacterium         Vibrio  #>      1.0000000      0.9666667      0.6666667      0.6000000      0.5000000  #> Ammopiptanthus  Mycobacterium       Medicago    Paucibacter   Enterococcus  #>      0.2000000      0.1666667      0.1333333      0.1000000      0.1000000 # Agglomerate by prevalence by genus altExp(mae[[2]], \"prev_genus\") <- agglomerateByPrevalence(     mae[[2]],     assay.type = \"relabundance\",     rank = \"Genus\",     prevalence = 0.2 / 100,     detection = 0.1 / 100 ) tse <- agglomerateByRank(mae[[2]], rank = \"Genus\") core <- getTop(tse, top = 19) # Get vector that tells whether taxa belongs to core taxa or not core_names <- rownames(tse) core_names[ !core_names %in% core ] <- \"other\" # Agglomerate based on core taxa rowData(tse)[[\"core\"]] <- core_names altExp(mae[[2]], \"core_genus\") <- agglomerateByVariable(   tse, by = \"rows\", group = \"core\") rowData(mae[[1]])[[\"sd\"]] <- rowSds(assay(mae[[1]], \"counts\"), na.rm = TRUE) hist(rowData(mae[[1]])[[\"sd\"]],     main = \"Standard deviation distribution of fatty acid abundances\",     xlab = \"Standard deviation\",     breaks = 20) mae[[1]] <- mae[[1]][ rowData(mae[[1]])[[\"sd\"]] > 0.5, ] relevant_fatty_acids <- c(   \"Docosahexaenoic acid 22:6n-3 (DHA)\",   \"Eicosapentaenoic acid 20:5n-3 (EPA)\",   \"Alpha-Linolenic acid 18:3n-3\",   \"Arachidonic acid 20:4n-6 (ARA)\",   \"Linoleic acid 18:2n-6\",   \"Oleic acid 18:1n-9\",   \"Palmitic acid 16:0\",   \"Stearic acid 18:0\" ) altExp(mae[[1]], \"relevant\") <- mae[[1]][     rownames(mae[[1]]) %in% relevant_fatty_acids, ]"},{"path":"/articles/case_study.html","id":"transformation","dir":"Articles","previous_headings":"Data preprocess","what":"Transformation","title":"HoloFoodR R package — case study","text":"transform metagenomic counts relative transformation centered log-ratio method tackle compositional data (see Quinn et al. (2019)). Fatty acid data already compositional measured concentration (mg/g). apply log10 transformation address skewness data. Finally, data standardized ensure features comparable scale.","code":"# Transform microbiome with centered log-ratio method mae[[2]] <- transformAssay(     mae[[2]],     assay.type = \"counts\",     method = \"relabundance\",     MARGIN = \"cols\" ) mae[[2]] <- transformAssay(     mae[[2]],     assay.type = \"counts\",     method = \"clr\",     pseudocount = TRUE,     MARGIN = \"cols\" ) #> A pseudocount of 0.5 was applied. #> A pseudocount of 0.5 was applied. #> A pseudocount of 0.5 was applied. mae[[1]] <- transformAssay(     mae[[1]],     assay.type = \"counts\",     method = \"log10\",     MARGIN = \"cols\" ) mae[[1]] <- transformAssay(     mae[[1]],     assay.type = \"log10\",     method = \"standardize\",     MARGIN = \"rows\" )"},{"path":[]},{"path":"/articles/case_study.html","id":"does-treatment-affect-fatty-acids","dir":"Articles","previous_headings":"Fatty acids, mass and treatment","what":"Does treatment affect fatty acids?","title":"HoloFoodR R package — case study","text":"aim investigate whether treatment associated changes fatty acid concentrations. begin, assess correlation treatment levels concentrations specific fatty acids observed, treatment appears significant effect fatty acid concentrations.","code":"# Get fatty acids that we are going to test tse <- altExp(mae[[1]], \"relevant\")  # For each fatty acid, test if the groups are different res <- lapply(rownames(tse), function(feat){     res <- pairwise.wilcox.test(       assay(tse, \"counts\")[feat, ], tse[[\"treatment_group\"]],       p.adjust.method = \"none\")     return(res) }) #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties # Combine results and adjust p -values res <- do.call(rbind, res) |> as.data.frame() res[[\"p.value\"]] <- p.adjust(res[[\"p.value\"]], method = \"fdr\") res |> datatable()"},{"path":"/articles/case_study.html","id":"is-there-association-between-salmon-size-and-fatty-acids","dir":"Articles","previous_headings":"Fatty acids, mass and treatment","what":"Is there association between salmon size and fatty acids?","title":"HoloFoodR R package — case study","text":"Additionally, worth exploring whether size salmon associated variations fatty acid levels. appears notable effect: larger salmon seem slightly higher concentrations fatty acids. trend can illustrated plots.  result initial analysis fatty acids, discovered size salmon appears association fatty acid concentrations, although association statistically significant.","code":"# Calculate correlation res <- getCrossAssociation(     mae, experiment1 = 1, experiment2 = 1, altexp1 = \"relevant\",     assay.type1 = \"standardize\", col.var2 = \"mass_scaled\",     test.signif = TRUE, method = \"pearson\" ) #> Calculating correlations... #> altexp1: relevant, altexp2: -, assay.type1: standardize, col.var1: -, assay.type2: -, col.var2: mass_scaled #> by: 1, function: stats::cor.test, method: pearson, test.signif: TRUE, p.adj.method: fdr, paired: FALSE, show.warnings: TRUE res #>                                  Var1        Var2       cor        pval #> 1  Docosahexaenoic acid 22:6n-3 (DHA) mass_scaled 0.4694512 0.008864359 #> 2 Eicosapentaenoic acid 20:5n-3 (EPA) mass_scaled 0.3666143 0.046294243 #> 3               Linoleic acid 18:2n-6 mass_scaled 0.3521072 0.056359869 #> 4                  Oleic acid 18:1n-9 mass_scaled 0.3585233 0.051714669 #> 5                  Palmitic acid 16:0 mass_scaled 0.4616248 0.010233011 #>        p_adj #> 1 0.02558253 #> 2 0.05635987 #> 3 0.05635987 #> 4 0.05635987 #> 5 0.02558253 p <- plotExpression(     altExp(mae[[1]], \"relevant\"),     features = rownames(altExp(mae[[1]], \"relevant\")),     x = \"mass_scaled\",     assay.type = \"standardize\",     scales = \"free\"     ) +   geom_smooth(method = \"lm\") p #> `geom_smooth()` using formula = 'y ~ x'"},{"path":[]},{"path":"/articles/case_study.html","id":"microbial-composition","dir":"Articles","previous_headings":"Microbiota, mass and treatment","what":"Microbial composition","title":"HoloFoodR R package — case study","text":"first step analysing microbiota data, summarize microbial composition relative abundance barplot.   Salmon gut seems dominated either genus Mycoplasma Photobacterium.","code":"p <- plotAbundance(     altExp(mae[[2]], \"core_genus\"),     assay.type = \"relabundance\",     col.var = \"treatment_group\",     facet.cols = TRUE, scales = \"free_x\"     ) p #> $abundance #>  #> $treatment_group"},{"path":"/articles/case_study.html","id":"is-alpha-diversity-associated-with-treatment-or-mass-of-salmon","dir":"Articles","previous_headings":"Microbiota, mass and treatment","what":"Is alpha diversity associated with treatment or mass of salmon?","title":"HoloFoodR R package — case study","text":"Next, let’s calculate Shannon alpha diversity index test whether diversity associated size treatment.  , observe potential effect, time treatment microbial diversity, although statistically significant. Additionally, association diversity mass. Let’s analyze whether can find similar effect beta diversity. perform Principal Coordinate Analysis (PCoA) Bray-Curtis dissimilarity.  seems handful samples differ others. However, effect seems associated treatment group.","code":"# Calculate alpha diversity mae[[2]] <- addAlpha(mae[[2]]) #> Warning: 'faith' index can be calculated only for TreeSE with rowTree(x) #> populated or with 'tree' provided separately.  # Plot with metadata p1 <- plotColData(mae[[2]], x = \"mass_scaled\", y = \"shannon_diversity\") +   geom_smooth(method = \"lm\") +   ggtitle(\"Shannon diversity association\\nwith body mass\") +   xlab(\"Scaled body mass\") + ylab(\"Shannon diversity index\") p2 <- plotColData(mae[[2]], x = \"treatment_group\", y = \"shannon_diversity\") +     geom_signif(comparisons = list(c(\"control\", \"treatment\")), ) +   ggtitle(\"Shannon diversity association\\nwith treatment\") +   xlab(\"Treatment group\") + ylab(\"Shannon diversity index\")  p1 + p2 #> `geom_smooth()` using formula = 'y ~ x' # Run PCoA mae[[2]] <- runMDS(     mae[[2]],     FUN = getDissimilarity,     method = \"bray\",     assay.type = \"relabundance\" )  # Display dissimilarity on a plot p <- plotReducedDim(mae[[2]], \"MDS\", colour_by = \"treatment_group\") +   ggtitle(\"PCoA plot of Bray-Curtis dissimilarity by treatment group\") p"},{"path":"/articles/case_study.html","id":"multi-omics-integration","dir":"Articles","previous_headings":"","what":"Multi-omics integration","title":"HoloFoodR R package — case study","text":"Multi-omic factor analysis (MOFA) (see Argelaguet et al. (2020)). allows us discover latent factors underlie biological differences taking consideration 2 omic assays. cite original authors, “MOFA can viewed statistically rigorous generalization (sparse) principal component analysis (PCA) multi-omics data”. applying MOFA analysis, goal determine whether metagenomics fatty acids exhibit shared variability, ultimately assessing whether microbial community associated fatty acid composition. Next, plot variances explained factor.  Factor 1 captures mostly variance metagenomics data. Factor 2 captures shared variability metagenomics fatty acid data. first factor captures highest amount variability metagenomics data, almost much variability remains captured second factor. exploring shared variability, first examine taxa’s variability captured Factor 1.  plot , can see first factor captures mostly variability Mycoplasma.  microbes, especially Photobacterium, positive association Factor 2. Many fatty acids also show significant weights, Linoleic acid 18:2n-6 sum omega-6 fatty acids positive weights, among others. indicates abundance Photobacterium increases, corresponding increase levels aforementioned fatty acids.","code":"mae_temp <- mae mae_temp[[2]] <- altExp(mae_temp[[2]], \"prev_genus\")  # Extract only transformed metagenomic assays for MOFA analysis assays(mae_temp[[1]]) <- assays(mae_temp[[1]])[     names(assays(mae_temp[[1]])) %in% c(\"standardize\") ] assays(mae_temp[[2]]) <- assays(mae_temp[[2]])[     names(assays(mae_temp[[2]])) %in% c(\"counts\") ]  # Transform MAE object to MOFA model model <- create_mofa_from_MultiAssayExperiment(mae_temp)  # Set model's options model_opts <- get_default_model_options(model) model_opts$num_factors <- 5 model_opts$likelihoods[[2]] <- \"poisson\" train_opts <- get_default_training_options(model) train_opts$maxiter <- 20000  # Change convergence mode to slightly improve accuracy train_opts$convergence_mode <- \"slow\"  # Prepare MOFA model model <- prepare_mofa(     object = model,     model_options = model_opts,     training_options = train_opts ) #> Checking data options... #> No data options specified, using default... #> Checking training options... #> Checking model options...  # Train model model <- run_mofa(model, use_basilisk = TRUE) #> Warning in run_mofa(model, use_basilisk = TRUE): No output filename provided. Using /tmp/RtmpGcJK0u/mofa_20241024-125203.hdf5 to store the trained model. #> Connecting to the mofapy2 package using basilisk.  #>     Set 'use_basilisk' to FALSE if you prefer to manually set the python binary using 'reticulate'. p <- plot_variance_explained(model) +     labs(title = \"Captured variance of fatty acid and metagenomic\\n\",         \"datasets across 5 factors\") +     theme(plot.title.position = \"plot\") p p1 <- plot_top_weights(model, view = 1, factors = 1, nfeatures = 25) +   labs(title = \"Fatty acids\") p2 <- plot_top_weights(model, view = 2, factors = 1, nfeatures = 25) +   labs(title = \"Microbiota\")  p1 + p2 p1 <- plot_top_weights(model, view = 1, factors = 2, nfeatures = 25) +   labs(title = \"Fatty acids\") p2 <- plot_top_weights(model, view = 2, factors = 2, nfeatures = 25) +   labs(title = \"Microbiota\")  p1 + p2"},{"path":"/articles/case_study.html","id":"conclusions","dir":"Articles","previous_headings":"","what":"Conclusions","title":"HoloFoodR R package — case study","text":"present case study demonstrated easy fast download large dataset transform data MultiAssayExperiment, turn gives researchers access extensive plethora downstream tools, mia MOFA2 can used pre-process visualize multi-omics data.","code":"sessionInfo() #> R version 4.4.1 (2024-06-14) #> Platform: x86_64-pc-linux-gnu #> Running under: Ubuntu 22.04.5 LTS #>  #> Matrix products: default #> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  #> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 #>  #> locale: #>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               #>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     #>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    #>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  #>  [9] LC_ADDRESS=C               LC_TELEPHONE=C             #> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        #>  #> time zone: UTC #> tzcode source: system (glibc) #>  #> attached base packages: #> [1] stats4    stats     graphics  grDevices utils     datasets  methods   #> [8] base      #>  #> other attached packages: #>  [1] scater_1.33.4                   scuttle_1.15.4                  #>  [3] reticulate_1.39.0               patchwork_1.3.0                 #>  [5] MOFA2_1.15.0                    miaViz_1.13.14                  #>  [7] ggraph_2.2.1                    ggplot2_3.5.1                   #>  [9] mia_1.13.47                     MGnifyR_1.1.0                   #> [11] HoloFoodR_0.99.10               TreeSummarizedExperiment_2.13.0 #> [13] Biostrings_2.73.2               XVector_0.45.0                  #> [15] SingleCellExperiment_1.27.2     MultiAssayExperiment_1.31.5     #> [17] SummarizedExperiment_1.35.4     Biobase_2.65.1                  #> [19] GenomicRanges_1.57.2            GenomeInfoDb_1.41.2             #> [21] IRanges_2.39.2                  S4Vectors_0.43.2                #> [23] BiocGenerics_0.51.3             MatrixGenerics_1.17.0           #> [25] matrixStats_1.4.1               ggsignif_0.6.4                  #> [27] DT_0.33                         dplyr_1.1.4                     #> [29] knitr_1.48                      BiocStyle_2.33.1                #>  #> loaded via a namespace (and not attached): #>   [1] splines_4.4.1               filelock_1.0.3              #>   [3] ggplotify_0.1.2             urltools_1.7.3              #>   [5] tibble_3.2.1                triebeard_0.4.1             #>   [7] polyclip_1.10-7             basilisk.utils_1.17.3       #>   [9] rpart_4.1.23                DirichletMultinomial_1.47.2 #>  [11] lifecycle_1.0.4             httr2_1.0.5                 #>  [13] lattice_0.22-6              MASS_7.3-61                 #>  [15] crosstalk_1.2.1             SnowballC_0.7.1             #>  [17] backports_1.5.0             magrittr_2.0.3              #>  [19] Hmisc_5.1-3                 sass_0.4.9                  #>  [21] rmarkdown_2.28              jquerylib_0.1.4             #>  [23] yaml_2.3.10                 RColorBrewer_1.1-3          #>  [25] cowplot_1.1.3               DBI_1.2.3                   #>  [27] minqa_1.2.8                 abind_1.4-8                 #>  [29] zlibbioc_1.51.2             Rtsne_0.17                  #>  [31] purrr_1.0.2                 yulab.utils_0.1.7           #>  [33] nnet_7.3-19                 tweenr_2.0.3                #>  [35] rappdirs_0.3.3              sandwich_3.1-1              #>  [37] GenomeInfoDbData_1.2.13     ggrepel_0.9.6               #>  [39] tokenizers_0.3.0            irlba_2.3.5.1               #>  [41] tidytree_0.4.6              pheatmap_1.0.12             #>  [43] vegan_2.6-8                 rbiom_1.0.3                 #>  [45] tidyjson_0.3.2              pkgdown_2.1.1               #>  [47] permute_0.9-7               DelayedMatrixStats_1.27.3   #>  [49] codetools_0.2-20            DelayedArray_0.31.14        #>  [51] ggforce_0.4.2               tidyselect_1.2.1            #>  [53] aplot_0.2.3                 UCSC.utils_1.1.0            #>  [55] farver_2.1.2                lme4_1.1-35.5               #>  [57] ScaledMatrix_1.13.0         viridis_0.6.5               #>  [59] base64enc_0.1-3             jsonlite_1.8.9              #>  [61] BiocNeighbors_1.99.3        decontam_1.25.0             #>  [63] tidygraph_1.3.1             Formula_1.2-5               #>  [65] systemfonts_1.1.0           ggnewscale_0.5.0            #>  [67] tools_4.4.1                 treeio_1.29.1               #>  [69] ragg_1.3.3                  Rcpp_1.0.13                 #>  [71] glue_1.8.0                  BiocBaseUtils_1.7.3         #>  [73] gridExtra_2.3               SparseArray_1.5.45          #>  [75] xfun_0.48                   mgcv_1.9-1                  #>  [77] HDF5Array_1.33.8            withr_3.0.1                 #>  [79] BiocManager_1.30.25         fastmap_1.2.0               #>  [81] basilisk_1.17.2             rhdf5filters_1.17.0         #>  [83] boot_1.3-31                 bluster_1.15.1              #>  [85] fansi_1.0.6                 digest_0.6.37               #>  [87] rsvd_1.0.5                  gridGraphics_0.5-1          #>  [89] R6_2.5.1                    textshaping_0.4.0           #>  [91] colorspace_2.1-1            lpSolve_5.6.21              #>  [93] UpSetR_1.4.0                utf8_1.2.4                  #>  [95] tidyr_1.3.1                 generics_0.1.3              #>  [97] data.table_1.16.2           DECIPHER_3.1.5              #>  [99] graphlayouts_1.2.0          httr_1.4.7                  #> [101] htmlwidgets_1.6.4           S4Arrays_1.5.11             #> [103] uwot_0.2.2                  pkgconfig_2.0.3             #> [105] gtable_0.3.5                janeaustenr_1.0.0           #> [107] htmltools_0.5.8.1           bookdown_0.41               #> [109] scales_1.3.0                png_0.1-8                   #> [111] corrplot_0.95               ggfun_0.1.6                 #> [113] rstudioapi_0.17.1           reshape2_1.4.4              #> [115] checkmate_2.3.2             nlme_3.1-166                #> [117] nloptr_2.1.1                rhdf5_2.49.0                #> [119] cachem_1.1.0                zoo_1.8-12                  #> [121] stringr_1.5.1               parallel_4.4.1              #> [123] vipor_0.4.7                 foreign_0.8-87              #> [125] desc_1.4.3                  pillar_1.9.0                #> [127] grid_4.4.1                  vctrs_0.6.5                 #> [129] slam_0.1-54                 BiocSingular_1.21.4         #> [131] beachmat_2.21.8             cluster_2.1.6               #> [133] beeswarm_0.4.0              htmlTable_2.4.3             #> [135] evaluate_1.0.1              mvtnorm_1.3-1               #> [137] cli_3.6.3                   compiler_4.4.1              #> [139] rlang_1.1.4                 crayon_1.5.3                #> [141] tidytext_0.4.2              labeling_0.4.3              #> [143] forcats_1.0.0               mediation_4.5.0             #> [145] plyr_1.8.9                  fs_1.6.4                    #> [147] ggbeeswarm_0.7.2            stringi_1.8.4               #> [149] viridisLite_0.4.2           BiocParallel_1.39.0         #> [151] assertthat_0.2.1            munsell_0.5.1               #> [153] lazyeval_0.2.2              Matrix_1.7-0                #> [155] dir.expiry_1.13.1           sparseMatrixStats_1.17.2    #> [157] Rhdf5lib_1.27.0             highr_0.11                  #> [159] memoise_2.0.1               igraph_2.1.1                #> [161] RcppParallel_5.1.9          bslib_0.8.0                 #> [163] ggtree_3.13.1               ape_5.8"},{"path":"/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Tuomas Borman. Author, maintainer. Artur Sannikov. Author. Leo Lahti. Author.","code":""},{"path":"/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Borman T, Sannikov , Lahti L (2024). HoloFoodR: R interface EBI HoloFood resource. R package version 0.99.10, https://github.com/EBI-Metagenomics/HoloFoodR.","code":"@Manual{,   title = {HoloFoodR: R interface to EBI HoloFood resource},   author = {Tuomas Borman and Artur Sannikov and Leo Lahti},   year = {2024},   note = {R package version 0.99.10},   url = {https://github.com/EBI-Metagenomics/HoloFoodR}, }"},{"path":"/index.html","id":"holofoodr-","dir":"","previous_headings":"","what":"R interface to EBI HoloFood resource","title":"R interface to EBI HoloFood resource","text":"R package developed streamlining integration analysis EMBL-EBI HoloFood data. utility package simplifies access resource, enabling direct loading data formats tailored multiomics downstream analytics. tool, users can efficiently search retrieve data EBI HoloFood resource. retrieved data available MultiAssayExperiment / TreeSummarizedExperiment format similarly data acquired MGnifyR package MGnify database. compatibility ensures users can seamlessly combine analyze datasets sources, leading valuable insights intricate biological systems. research received funding Horizon 2020 Programme European Union within FindingPheno project grant agreement 952914.","code":""},{"path":"/index.html","id":"findingpheno-","dir":"","previous_headings":"","what":"FindingPheno","title":"R interface to EBI HoloFood resource","text":"FindingPheno, EU-funded project, dedicated developing computational tools methodologies integration analysis multi-omics data. primary objective deepen understanding interactions hosts microbiomes. FindingPheno website","code":""},{"path":"/index.html","id":"holofood-","dir":"","previous_headings":"","what":"HoloFood","title":"R interface to EBI HoloFood resource","text":"HoloFood, project funded EU’s Horizon 2020 programme, employed holistic, “hologenomic”, approach enhance efficiency food production systems. involved exploring biomolecular physiological processes triggered incorporation feed additives novel sustainable feeds farmed animals. HoloFood database, hosted European Bioinformatics Institute (EMBL-EBI), houses data gathered project, encompassing multiple omics, including metabolomics various biomolecular measurements. Notably, include data metagenomic untargeted metabolomic analyses. However, metagenomic data project can accessed MGnify database, untargeted metabolomic data stored MetaboLights database. explore available datasets HoloFood, can utilize API browser. HoloFood website HoloFood publications API browser API documentation","code":""},{"path":"/index.html","id":"mgnify-","dir":"","previous_headings":"","what":"MGnify","title":"R interface to EBI HoloFood resource","text":"EMBL-EBI’s MGnify serves repository microbiome data, offering wide array analyses encompassing metabarcoding, metatranscriptomic, metagenomic datasets diverse environments. platform provides comprehensive taxonomic assignments functional annotations datasets. data can accessed MGnifyR package. MGnify website MGnifyR","code":""},{"path":"/index.html","id":"metabolights-","dir":"","previous_headings":"","what":"MetaboLights","title":"R interface to EBI HoloFood resource","text":"MetaboLights, managed EMBL-EBI, serves repository metabolomic data. can accessed HoloFoodR package. MetaboLights website","code":""},{"path":"/index.html","id":"miaverse-","dir":"","previous_headings":"","what":"miaverse","title":"R interface to EBI HoloFood resource","text":"miaverse R/Bioconductor framework specialized microbiome downstream data analysis, leveraging TreeSummarizedExperiment class. offers comprehensive suite tools microbiome bioinformatics. Additionally, miaverse includes tutorial book Orchestrating Microbiome Analysis (OMA), aims guide users conducting microbiome data analysis sharing best practices microbiome data science. miaverse website Orchestrating Microbiome Analysis (OMA)","code":""},{"path":[]},{"path":"/index.html","id":"bioc-release","dir":"","previous_headings":"Installation","what":"Bioc-release","title":"R interface to EBI HoloFood resource","text":"","code":"if (!requireNamespace(\"BiocManager\", quietly = TRUE))     install.packages(\"BiocManager\")  BiocManager::install(\"HoloFoodR\")"},{"path":"/index.html","id":"bioc-devel","dir":"","previous_headings":"Installation","what":"Bioc-devel","title":"R interface to EBI HoloFood resource","text":"","code":"if (!requireNamespace(\"BiocManager\", quietly = TRUE))     install.packages(\"BiocManager\")  # The following initializes usage of Bioc devel BiocManager::install(version=\"devel\")  BiocManager::install(\"HoloFoodR\")"},{"path":"/index.html","id":"github","dir":"","previous_headings":"Installation","what":"GitHub","title":"R interface to EBI HoloFood resource","text":"","code":"remotes::install_github(\"EBI-Metagenomics/HoloFoodR\")"},{"path":"/index.html","id":"basic-usage","dir":"","previous_headings":"","what":"Basic usage","title":"R interface to EBI HoloFood resource","text":"detailed instructions read associated function help, function reference page vignette (vignette(\"HoloFoodR\"))","code":"library(HoloFoodR)  # Search samples samples <- doQuery(\"samples\")  # Search animals animals <- doQuery(\"animal\")  # Fetch data on certain sample samples <- c(\"ACCESSION_ID\") sample_data <- getData(accession.type = \"samples\", accession = samples)  # Fetch data on genome catalogues genome_catalogues <- getData(type = \"genome-catalogues\")  # Fetch data on genomes in certain genome catalogue catalogues <- c(\"ACCESSION_ID\") genomes <- getData(     type = \"genomes\",     accession.type = \"genome-catalogues\",     accession = catalogues)  # Fetch data on untargeted metabolites metabolites <- getMetaboLights(study_id)  # Fetch data as MultiAssayExperiment samples <- c(\"ACCESSION_ID\") mae <- getResult(accession = samples)"},{"path":"/reference/HoloFoodR.html","id":null,"dir":"Reference","previous_headings":"","what":"HoloFoodR package — HoloFoodR","title":"HoloFoodR package — HoloFoodR","text":"HoloFoodR implements interface EBI HoloFood database. See vignette general introduction package, [HoloFood](https://www.holofood.eu/) general HoloFood information, [API documentation](https://docs.holofooddata.org/api.html) details JSONAPI implementation.","code":""},{"path":[]},{"path":"/reference/HoloFoodR.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"HoloFoodR package — HoloFoodR","text":"Maintainer: Tuomas Borman tuomas.v.borman@utu.fi (ORCID) Authors: Artur Sannikov arsann@utu.fi (ORCID) Leo Lahti leo.lahti@iki.fi (ORCID)","code":""},{"path":"/reference/addMGnify.html","id":null,"dir":"Reference","previous_headings":"","what":"Add results from MGnifyR to HoloFoodR results — addMGnify","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"Add results MGnifyR HoloFoodR results","code":""},{"path":"/reference/addMGnify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"","code":"addMGnify(x, y, ...)  # S4 method for class 'SummarizedExperiment,MultiAssayExperiment' addMGnify(   x,   y,   exp.name1 = \"metagenomic\",   exp.name2 = \"metagenomic_amplicon\",   replace = TRUE,   ... )  # S4 method for class 'SummarizedExperiment,SummarizedExperiment' addMGnify(x, y, ...)  # S4 method for class 'SummarizedExperiment,ANY' addMGnify(x, y, id.col1 = \"sample_biosample\", id.col2 = \"accession\", ...)"},{"path":"/reference/addMGnify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"x SummarizedExperiment. Results MGnifyR::getResult(). y MultiAssayExperiment SummarizedExperiment data.frame-like table. Results HoloFoodR::getResult() sample metadata . ... optional arguments used currently. exp.name1 Character scalar. Specifies name experiment added y. (Default: \"metagenomic\") exp.name2 Character scalar. Specifies name experiment HoloFoodR results. experiment used match IDs MGnify data. (Default: \"metagenomic_amplicon\") replace Logical scalar. Whether replace template experiment. (Default: TRUE) id.col1 Character scalar. Specifies name column colData(x) includes HoloFood identifiers. (Default: \"sample_biosample\") id.col2 Character scalar. Specifies name column colData(y[[exp.name2]]) includes HoloFood identifiers. (Default: \"accession\")","code":""},{"path":"/reference/addMGnify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"MultiAssayExperiment","code":""},{"path":"/reference/addMGnify.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"Metagenomic data found MGnify rather HoloFoodR, two databases use different sample identifiers. However, MGnify's sample metadata includes references identifiers used HoloFood database, making straightforward convert sample IDs alignment HoloFood data. Despite , HoloFood contains additional metadata available MGnify. Moreover, integrating data MultiAssayExperiment maintaining accurate sample system matches can challenging. function designed simplify tasks, enabling seamless integration MGnify data HoloFood data retrieval database. need input returned data MGnifyR::getResult() HoloFoodR::getResult() functions.","code":""},{"path":"/reference/addMGnify.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"","code":"if (FALSE) { # \\dontrun{ # Get data from HoloFood database mae <- HoloFoodR::getResult(     salmon_sample_ids,     use.cache = TRUE )  # Get data from MGnify database mg <- MgnifyClient(     useCache = TRUE,     cacheDir = \".MGnifyR_cache\" ) tse <- MGnifyR::getResult(     mg,     accession = mgnify_analyses_ids,     get.func = FALSE )  # Add MGnify data to HoloFood data mae <- addMGnify(tse, mae) } # }"},{"path":"/reference/doQuery.html","id":null,"dir":"Reference","previous_headings":"","what":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"Search HoloFood database animals, genome catalogues, samples, viral catalogues","code":""},{"path":"/reference/doQuery.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"","code":"doQuery(type, flatten = TRUE, ...)"},{"path":"/reference/doQuery.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"type Character scalar specifying type data query. Must one following options: \"animals\", \"genome-catalogues\", \"samples\" \"viral-catalogues\". flatten Logical scalar specifying whether flatten resulting data.frame. means columns multiple values separated multiple columns. (Default: TRUE) ... optional arguments: max.hits NULL integer scalar specifying   maximum number results fetch. NULL, results fetched.   (Default: NULL) spread.sample.types Logical scalar specifying whether   create spread sample types column animals data. animals data,   sample types column might multiple values might hard   explore. argument specifies whether create presence/absence table   sample types. (Default: TRUE) use.cache Logical scalar specifying whether   use cache. (Default: FALSE) cache.dir Character scalar specifying cache   directory. (Default: tempdir()) clear.cache Logical scalar specifying whether   remove clear cache (Default: FALSE)","code":""},{"path":"/reference/doQuery.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"data.frame","code":""},{"path":"/reference/doQuery.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"doQuery flexible query function can utilized search available animals, genome catalogues, samples, viral catalogues. Search results can filtered; example, animals can filtered based available samples. See [Api browser](https://www.holofooddata.org/api/docs) information filters. can find help customizing queries [](https://emg-docs.readthedocs.io/en/latest/api.html#customising-queries).","code":""},{"path":"/reference/doQuery.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"","code":"# Find animals results. The maximum amount of results is 100. Use filter # so that only chicken is searched. res <- doQuery(\"animals\", max.hits = 100, system = \"chicken\") head(res) #>        accession  system                                   canonical_url #> 1 SAMEA112904733 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904733 #> 2 SAMEA112904734 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904734 #> 3 SAMEA112904735 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904735 #> 4 SAMEA112904736 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904736 #> 5 SAMEA112904737 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904737 #> 6 SAMEA112904738 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904738 #>   histological host_genomic inflammatory_markers metabolomic #> 1        FALSE         TRUE                FALSE       FALSE #> 2         TRUE         TRUE                 TRUE       FALSE #> 3         TRUE         TRUE                 TRUE       FALSE #> 4        FALSE         TRUE                FALSE       FALSE #> 5         TRUE         TRUE                 TRUE       FALSE #> 6        FALSE         TRUE                FALSE       FALSE #>   metabolomic_targeted metagenomic_assembly metatranscriptomic transcriptomic #> 1                FALSE                 TRUE              FALSE          FALSE #> 2                 TRUE                 TRUE               TRUE           TRUE #> 3                 TRUE                 TRUE              FALSE           TRUE #> 4                FALSE                 TRUE              FALSE          FALSE #> 5                 TRUE                 TRUE               TRUE           TRUE #> 6                 TRUE                 TRUE              FALSE          FALSE"},{"path":"/reference/getData.html","id":null,"dir":"Reference","previous_headings":"","what":"Get data from HoloFood database — getData","title":"Get data from HoloFood database — getData","text":"Get data HoloFood database","code":""},{"path":"/reference/getData.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get data from HoloFood database — getData","text":"","code":"getData(   type = NULL,   accession.type = NULL,   accession = NULL,   flatten = FALSE,   ... )"},{"path":"/reference/getData.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get data from HoloFood database — getData","text":"type NULL character scalar specifying type data query. Must one following options: \"analysis-summaries\", \"animals\", \"genome-catalogues\", \"samples\", \"sample_metadata_markers\" \"viral-catalogues\". genome viral catalogues fetched accession ID, type can also \"genomes\" \"fragments\". (Default: NULL) accession.type NULL character scalar specifying type accession IDs. Must one following options: \"animals\", \"genome-catalogues\", \"samples\" \"viral-catalogues\". (Default: NULL) accession NULL character vector specifying accession IDs type accession.type. (Default: NULL) flatten Logical scalar specifying whether flatten resulting data.frame. means columns multiple values separated multiple columns. (Default: FALSE) ... optional arguments: max.hits NULL integer scalar specifying   maximum number results fetch. NULL, results fetched.   (Default: NULL) use.cache Logical scalar specifying whether   use cache (Default: FALSE) cache.dir Character scalar specifying cache   directory. (Default: tempdir()) clear.cache Logical scalar specifying whether   remove clear cache (Default: FALSE)","code":""},{"path":"/reference/getData.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get data from HoloFood database — getData","text":"list data.frame","code":""},{"path":"/reference/getData.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get data from HoloFood database — getData","text":"getData, can fetch data database. Compared getResult, function flexible since can fetch kind data database. However, function returns data without wrangling list data.frame optimized format fetching data samples. Search results can filtered; example, animals can filtered based available samples. See [Api browser](https://www.holofooddata.org/api/docs) information filters. can find help customizing queries [](https://emg-docs.readthedocs.io/en/latest/api.html#customising-queries).","code":""},{"path":[]},{"path":"/reference/getData.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get data from HoloFood database — getData","text":"","code":"# Find genome catalogues catalogues <- getData(type = \"genome-catalogues\") head(catalogues) #>                 id                   title #> 1  salmon-gut-v2-0  HoloFood Salmon Gut v2 #> 2 chicken-gut-v2-0 HoloFood Chicken Gut v2 #>                                         biome related_mag_catalogue_id  system #> 1  root:Host-associated:Fish:Digestive system  non-model-fish-gut-v2-0  salmon #> 2 root:Host-associated:Birds:Digestive system       chicken-gut-v1-0-1 chicken #>   analysis_summaries #> 1       c(\"HoloF.... #> 2       c(\"HoloF....  # Find genomes based on certain genome catalogue iD res <- getData(     type = \"genomes\", accession.type = \"genome-catalogues\",     accession = catalogues[1, \"id\"], max.hits = 100) # See the data. head(res) #>       accession cluster_representative #> 1 MGYG000307500          MGYG000307500 #> 2 MGYG000307501          MGYG000307501 #> 3 MGYG000307502          MGYG000307501 #> 4 MGYG000307503          MGYG000299622 #> 5 MGYG000307504          MGYG000299579 #> 6 MGYG000307505          MGYG000299579 #>                                                                                                                      taxonomy #> 1                                                    Bacteria > Firmicutes_A > Clostridia > Oscillospirales > Ruminococcaceae #> 2 Bacteria > Proteobacteria > Gammaproteobacteria > Pseudomonadales > Pseudomonadaceae > Pseudomonas > Pseudomonas aeruginosa #> 3 Bacteria > Proteobacteria > Gammaproteobacteria > Pseudomonadales > Pseudomonadaceae > Pseudomonas > Pseudomonas aeruginosa #> 4          Bacteria > Proteobacteria > Gammaproteobacteria > Enterobacterales > Aeromonadaceae > Aeromonas > Aeromonas sobria #> 5                                                       Bacteria > Firmicutes > Bacilli > Mycoplasmatales > Mycoplasmoidaceae #> 6                                                       Bacteria > Firmicutes > Bacilli > Mycoplasmatales > Mycoplasmoidaceae #>                                                representative_url metadata1 #> 1 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000307500       196 #> 2 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000307501       197 #> 3 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000307501       198 #> 4 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000299622       199 #> 5 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000299579       200 #> 6 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000299579       201 #>   metadata.Genome_type metadata.Length metadata.N_contigs metadata.N50 #> 1                  MAG         1319783                 19       153761 #> 2                  MAG         6676386                136        83355 #> 3                  MAG         4766025               4741         1104 #> 4                  MAG         2381842                367         8519 #> 5                  MAG          681326                 38        37281 #> 6                  MAG          627706                 26        43119 #>   metadata.GC_content metadata.Completeness metadata.Contamination #> 1                27.3                 85.79                    0.0 #> 2               66.27                 98.37                  0.192 #> 3               64.82                 68.36                  2.053 #> 4               59.04                 58.95                   0.09 #> 5               25.14                 96.02                  0.384 #> 6               25.04                 95.25                  0.384 #>   metadata.rRNA_5S metadata.rRNA_16S metadata.rRNA_23S metadata.tRNAs #> 1              0.0               0.0               0.0             18 #> 2              0.0               0.0               0.0             17 #> 3            93.28              36.2             99.04             15 #> 4             91.6               0.0               0.0             14 #> 5              0.0             99.74             99.59             19 #> 6              0.0             99.74             99.59             19 #>   metadata.Genome_accession metadata.Sample_accession metadata.Study_accession #> 1               ERZ15182294            SAMEA112246717                ERP136460 #> 2               ERZ15182298            SAMEA112246686                ERP136460 #> 3               ERZ15182309            SAMEA112246683                ERP136460 #> 4               ERZ15182314            SAMEA112246713                ERP136460 #> 5               ERZ15233654            SAMEA112264417                ERP125469 #> 6               ERZ15233655            SAMEA112264472                ERP125469 #>   metadata.Country metadata.Continent #> 1           Norway             Europe #> 2           Norway             Europe #> 3           Norway             Europe #> 4           Norway             Europe #> 5           Norway             Europe #> 6           Norway             Europe #>                                                                                                                                       metadata.FTP_download #> 1 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0003075/MGYG000307500/genomes1/MGYG000307500.gff.gz #> 2 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0003075/MGYG000307501/genomes1/MGYG000307501.gff.gz #> 3 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0003075/MGYG000307501/genomes1/MGYG000307502.gff.gz #> 4 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0002996/MGYG000299622/genomes1/MGYG000307503.gff.gz #> 5 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0002995/MGYG000299579/genomes1/MGYG000307504.gff.gz #> 6 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0002995/MGYG000299579/genomes1/MGYG000307505.gff.gz #>   metadata.Host_species annotations.cazy.GH annotations.cazy.PL #> 1           Salmo salar                   7                   0 #> 2           Salmo salar                  24                   0 #> 3           Salmo salar                  24                   0 #> 4           Salmo salar                  27                   0 #> 5           Salmo salar                   0                   0 #> 6           Salmo salar                   0                   0 #>   annotations.cazy.CE annotations.cazy.AA annotations.cazy.CB #> 1                   0                   0                   1 #> 2                   3                   1                   4 #> 3                   3                   1                   4 #> 4                   1                   1                   6 #> 5                   0                   0                   0 #> 6                   0                   0                   0 #>   annotations.cazy.GT annotations.cazy.CL #> 1                   7                   0 #> 2                  34                   0 #> 3                  34                   0 #> 4                  20                   0 #> 5                   2                   0 #> 6                   2                   0 # It includes for instance summary of the CAZy # (Carbohydrate-Active enZymes) annotations as a counts per category cazy <- res[ , grepl(\"annotations.cazy\", colnames(res)), drop = FALSE] head(cazy) #>   annotations.cazy.GH annotations.cazy.PL annotations.cazy.CE #> 1                   7                   0                   0 #> 2                  24                   0                   3 #> 3                  24                   0                   3 #> 4                  27                   0                   1 #> 5                   0                   0                   0 #> 6                   0                   0                   0 #>   annotations.cazy.AA annotations.cazy.CB annotations.cazy.GT #> 1                   0                   1                   7 #> 2                   1                   4                  34 #> 3                   1                   4                  34 #> 4                   1                   6                  20 #> 5                   0                   0                   2 #> 6                   0                   0                   2 #>   annotations.cazy.CL #> 1                   0 #> 2                   0 #> 3                   0 #> 4                   0 #> 5                   0 #> 6                   0 # Moreover, it includes a sample list. This sample list represents a  # collection of samples where the MAG was identified. Thr data has also the # completeness of MAG in a sample. head(res[ c(\"metadata.Sample_accession\", \"metadata.Completeness\")]) #>   metadata.Sample_accession metadata.Completeness #> 1            SAMEA112246717                 85.79 #> 2            SAMEA112246686                 98.37 #> 3            SAMEA112246683                 68.36 #> 4            SAMEA112246713                 58.95 #> 5            SAMEA112264417                 96.02 #> 6            SAMEA112264472                 95.25"},{"path":"/reference/getMetaboLights.html","id":null,"dir":"Reference","previous_headings":"","what":"Get metabolomic data from MetaboLights database — getMetaboLights","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"Get metabolomic data MetaboLights database","code":""},{"path":"/reference/getMetaboLights.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"","code":"getMetaboLights(study.id, ...)  getMetaboLightsFile(study.id, file, ...)"},{"path":"/reference/getMetaboLights.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"study.id character vector specifying study identifier data going fetched MetaboLights database. ... optional arguments: cache.dir Character scalar specifying directory   downloaded file stored. (Default: tempdir()) timeout Integer scalar specifying timeout   seconds loading file. (Default: 5*60) file character vector specifying files fetched.","code":""},{"path":"/reference/getMetaboLights.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"list","code":""},{"path":"/reference/getMetaboLights.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"HoloFood database primarily comprises targeted metabolomic data, omitting non-targeted metabolomic information. Nonetheless, features URLs linking studies within MetaboLights database. functionality enables users access non-targeted metabolomic data. getMetaboLights function returns structured list encompassing processed data data.frame format study metadata, assay metadata, assay. metadata includes file names spectra data. files can loaded getMetaboLightsFile. Alternatively, identified study files fetch, can refer [vignette](https://rformassspectrometry.github.io/MsIO/articles/MsIO.html#loading-data--metabolights) instructions loading data directly MsExperiment object, specifically designed metabolomics spectra data.","code":""},{"path":[]},{"path":"/reference/getMetaboLights.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"","code":"# This example is not run, because the server fails to respond sometimes. if( FALSE ){     res <- getMetaboLights(\"MTBLS4381\")     file_paths <- getMetaLightsFile(         study.id = \"MTBLS4381\",         file = res[[\"assay_meta\"]][[\"Raw Spectral Data File\"]]         ) }"},{"path":"/reference/getResult.html","id":null,"dir":"Reference","previous_headings":"","what":"Get data on samples from HoloFood database — getResult","title":"Get data on samples from HoloFood database — getResult","text":"Get data samples HoloFood database","code":""},{"path":"/reference/getResult.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get data on samples from HoloFood database — getResult","text":"","code":"getResult(accession, ...)"},{"path":"/reference/getResult.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get data on samples from HoloFood database — getResult","text":"accession Character vector specifying accession IDs type samples. ... optional arguments: use.cache Logical scalar specifying whether   use cache. Note get.metabolomic = TRUE specified,   file MetaboLights stored local system location   specified cache.dir despite value use.cache.   (Default: FALSE) cache.dir Character scalar specifying cache   directory. (Default: tempdir()) clear.cache Logical scalar specifying whether   use.cache (Default: FALSE) assay.type Character scalar specifying name   assay resulting TreeSummarizedExperiment object.   (Default: \"counts\") get.metabolomic Logical scalar specifying whether   retrieve processed metabolomic data MetaboLights database.   retrieving spectra data, refer   getMetaboLights documentation.   (Default: FALSE)","code":""},{"path":"/reference/getResult.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get data on samples from HoloFood database — getResult","text":"MultiAssayExperiment","code":""},{"path":"/reference/getResult.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get data on samples from HoloFood database — getResult","text":"getResult, can fetch data samples HoloFood database. Compared getData, function convenient fetching samples data converts data MultiAssayExperiment different omics stored TreeSummarizedExperiment objects optimized downstream analytics. Columns returned MultiAssayExperiment individual animals. columns linked individual samples stored TreeSummarizedExperiment objects. HoloFood database lacks non-targeted metabolomic data can fetched MetaboLights resource. Certain datasets include processed features. datasets can retrieved function getResult integrates metabolomic data datasets HoloFood. Furthermore, HoloFoodR database include metagenomic assembly data, users can access data MGnify database. MGnifyR package provides convenient interface accessing database. employing MGnifyR::getResult(), users can obtain data formatted MultiAssayExperiment object, containing multiple TreeSummarizedExperiment objects. Consequently, data HoloFood MGnify databases inherently compatible subsequent downstream analysis.","code":""},{"path":[]},{"path":"/reference/getResult.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get data on samples from HoloFood database — getResult","text":"","code":"# Find samples on certain animal samples <- doQuery(\"samples\", animal_accession = \"SAMEA112904746\")  # Get the data mae <- getResult(samples[[\"accession\"]]) #> Warning: Data for the following samples cannot be found. The sample types are metagenomic_assembly, host_genomic, transcriptomic and metatranscriptomic. (Note that metagenomic assemblies can be found from the MGnify database. See MGnifyR package.): #> 'SAMEA10130039', 'SAMEA13389496', 'SAMEA13389497', 'SAMEA13901618', 'SAMEA13901619', 'SAMEA13929785', 'SAMEA7571815' mae #> A MultiAssayExperiment object of 8 listed #>  experiments with user-defined names and respective classes. #>  Containing an ExperimentList class object of length 8: #>  [1] BIOGENIC AMINES: TreeSummarizedExperiment with 7 rows and 2 columns #>  [2] FATTY ACIDS: TreeSummarizedExperiment with 19 rows and 2 columns #>  [3] HISTOLOGY: TreeSummarizedExperiment with 20 rows and 2 columns #>  [4] INFLAMMATORY MARKERS: TreeSummarizedExperiment with 14 rows and 2 columns #>  [5] metagenomic_assembly: TreeSummarizedExperiment with 0 rows and 2 columns #>  [6] host_genomic: TreeSummarizedExperiment with 0 rows and 2 columns #>  [7] transcriptomic: TreeSummarizedExperiment with 0 rows and 2 columns #>  [8] metatranscriptomic: TreeSummarizedExperiment with 0 rows and 1 columns #> Functionality: #>  experiments() - obtain the ExperimentList instance #>  colData() - the primary/phenotype DataFrame #>  sampleMap() - the sample coordination DataFrame #>  `$`, `[`, `[[` - extract colData columns, subset, or experiment #>  *Format() - convert into a long or wide DataFrame #>  assays() - convert ExperimentList to a SimpleList of matrices #>  exportClass() - save data to flat files"}]
+[{"path":"/articles/HoloFoodR.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFoodR package designed ease access EBI’s HoloFoodR resource, allowing searching retrieval multiple datasets downstream analysis. HoloFood database encompass metagenomics data; however, data stored within MGnify database. packages offer analogous functionalities, streamlining integration data enhancing accessibility. TreeSummarizedExperiment","code":""},{"path":"/articles/HoloFoodR.html","id":"installation","dir":"Articles","previous_headings":"","what":"Installation","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFoodR hosted Bioconductor, can installed using via BiocManager.","code":"BiocManager::install(\"HoloFoodR\")"},{"path":"/articles/HoloFoodR.html","id":"load-the-package","dir":"Articles","previous_headings":"","what":"Load the package","title":"HoloFoodR: interface to HoloFoodR database","text":"installed, HoloFoodR made available usual way.","code":"library(HoloFoodR) #> Loading required package: MultiAssayExperiment #> Loading required package: SummarizedExperiment #> Loading required package: MatrixGenerics #> Loading required package: matrixStats #>  #> Attaching package: 'MatrixGenerics' #> The following objects are masked from 'package:matrixStats': #>  #>     colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse, #>     colCounts, colCummaxs, colCummins, colCumprods, colCumsums, #>     colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs, #>     colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats, #>     colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds, #>     colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads, #>     colWeightedMeans, colWeightedMedians, colWeightedSds, #>     colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet, #>     rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods, #>     rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps, #>     rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins, #>     rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks, #>     rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars, #>     rowWeightedMads, rowWeightedMeans, rowWeightedMedians, #>     rowWeightedSds, rowWeightedVars #> Loading required package: GenomicRanges #> Loading required package: stats4 #> Loading required package: BiocGenerics #>  #> Attaching package: 'BiocGenerics' #> The following objects are masked from 'package:stats': #>  #>     IQR, mad, sd, var, xtabs #> The following objects are masked from 'package:base': #>  #>     anyDuplicated, aperm, append, as.data.frame, basename, cbind, #>     colnames, dirname, do.call, duplicated, eval, evalq, Filter, Find, #>     get, grep, grepl, intersect, is.unsorted, lapply, Map, mapply, #>     match, mget, order, paste, pmax, pmax.int, pmin, pmin.int, #>     Position, rank, rbind, Reduce, rownames, sapply, saveRDS, setdiff, #>     table, tapply, union, unique, unsplit, which.max, which.min #> Loading required package: S4Vectors #>  #> Attaching package: 'S4Vectors' #> The following object is masked from 'package:utils': #>  #>     findMatches #> The following objects are masked from 'package:base': #>  #>     expand.grid, I, unname #> Loading required package: IRanges #> Loading required package: GenomeInfoDb #> Loading required package: Biobase #> Welcome to Bioconductor #>  #>     Vignettes contain introductory material; view with #>     'browseVignettes()'. To cite Bioconductor, see #>     'citation(\"Biobase\")', and for packages 'citation(\"pkgname\")'. #>  #> Attaching package: 'Biobase' #> The following object is masked from 'package:MatrixGenerics': #>  #>     rowMedians #> The following objects are masked from 'package:matrixStats': #>  #>     anyMissing, rowMedians #> Loading required package: TreeSummarizedExperiment #> Loading required package: SingleCellExperiment #> Loading required package: Biostrings #> Loading required package: XVector #>  #> Attaching package: 'Biostrings' #> The following object is masked from 'package:base': #>  #>     strsplit"},{"path":"/articles/HoloFoodR.html","id":"functionalities","dir":"Articles","previous_headings":"","what":"Functionalities","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFoodR offers three functions doQuery(), getResult() getData() can utilized search fetch data HoloFood database. tutorial, demonstrate search animals, subset animals based whether specific sample type, finally fetch data samples. Note can done doQuery() getResult() (getData() getResult()) utilizing query filters. tutorial demonstrating functionality package. Additionally, package includes getMetaboLights() function can utilized retrieve metabolomic data MetaboLights database.","code":""},{"path":"/articles/HoloFoodR.html","id":"search-data","dir":"Articles","previous_headings":"Functionalities","what":"Search data","title":"HoloFoodR: interface to HoloFoodR database","text":"search animals, genome catalogues, samples viral catalogues, can use doQuery() function. can also use getData() doQuery() optimized searching datatypes. example, instead nested list sample types doQuery() returns sample types presence/absence table convenient. search animals, subset based whether include histological samples. Note can done also using query filters. doQuery() returns data.frame including information type data searched.","code":"animals <- doQuery(\"animals\", max.hits = 100) animals <- animals[ animals[[\"histological\"]], ]  colnames(animals) |> head() #> [1] \"accession\"            \"system\"               \"canonical_url\"        #> [4] \"histological\"         \"host_genomic\"         \"inflammatory_markers\""},{"path":"/articles/HoloFoodR.html","id":"get-data","dir":"Articles","previous_headings":"Functionalities","what":"Get data","title":"HoloFoodR: interface to HoloFoodR database","text":"Now information animal histological samples. Let’s get data animals know sample IDs fetch. returned value getData() function list. can data also data.frame specify flatten = TRUE. data information animals including samples drawn . elements list data.frames. example, “samples” table contains information samples drawn animals specified input. Now can collect sample IDs.","code":"animal_data <- getData(     accession.type = \"animals\", accession = animals[[\"accession\"]]) samples <- animal_data[[\"samples\"]]  colnames(samples) |> head() #> [1] \"accession\"        \"title\"            \"sample_type\"      \"animal\"           #> [5] \"canonical_url\"    \"metagenomics_url\" sample_ids <- unique(samples[[\"accession\"]])"},{"path":"/articles/HoloFoodR.html","id":"get-data-on-samples","dir":"Articles","previous_headings":"Functionalities","what":"Get data on samples","title":"HoloFoodR: interface to HoloFoodR database","text":"get data samples, can utilize getResult() function. returns data MultiAssayExperiment (MAE) format. MAE object stores individual omics TreeSummarizedExperiment (TreeSE) objects. TreeSE, column represents sample rows represent features.","code":"mae <- getResult(sample_ids) #> Warning: Data for the following samples cannot be found. The sample types are metagenomic_assembly, host_genomic, transcriptomic and metatranscriptomic. (Note that metagenomic assemblies can be found from the MGnify database. See MGnifyR package.): #> 'SAMEA10130025', 'SAMEA13389405', 'SAMEA13389406', 'SAMEA13901590', 'SAMEA13901591', 'SAMEA13929779', 'SAMEA7697591', 'SAMEA10130091', 'SAMEA13389692', 'SAMEA13389693', 'SAMEA13901708', 'SAMEA7571845', 'SAMEA10158030', 'SAMEA13389419', 'SAMEA13389420', 'SAMEA13901594', 'SAMEA13901595', 'SAMEA13929781', 'SAMEA7697592', 'SAMEA10130039', 'SAMEA13389496', 'SAMEA13389497', 'SAMEA13901618', 'SAMEA13901619', 'SAMEA13929785', 'SAMEA7571815', 'SAMEA10130112', 'SAMEA13389794', 'SAMEA13389795', 'SAMEA13901758', 'SAMEA13901759', 'SAMEA13929811', 'SAMEA7571864', 'SAMEA10158022', 'SAMEA13389146', 'SAMEA13389147', 'SAMEA13901511', 'SAMEA13901512', 'SAMEA13929767', 'SAMEA7571777', 'SAMEA10130019', 'SAMEA13389353', 'SAMEA13389354', 'SAMEA13389355', 'SAMEA13901574', 'SAMEA13901575', 'SAMEA14095991', 'SAMEA7722475', 'SAMEA10130101', 'SAMEA13389738', 'SAMEA13389739', 'SAMEA13901730', 'SAMEA13901731', 'SAMEA13929802', 'SAMEA7571856', 'SAMEA10455480', 'SAMEA13389220', 'SAMEA10129993', 'SAMEA13389183', 'SAMEA13389184', 'SAMEA13901520', 'SAMEA13901521', 'SAMEA7697579', 'SAMEA10130017', 'SAMEA13389345', 'SAMEA13389346', 'SAMEA13901571', 'SAMEA13901572', 'SAMEA13929772', 'SAMEA7571801', 'SAMEA10130113', 'SAMEA13389807', 'SAMEA13389808', 'SAMEA13901762', 'SAMEA13901763', 'SAMEA13929813', 'SAMEA7571866', 'SAMEA10455481', 'SAMEA13389227', 'SAMEA10455479', 'SAMEA13389169', 'SAMEA10130020', 'SAMEA13389357', 'SAMEA13389358', 'SAMEA13901576', 'SAMEA13901577', 'SAMEA13929773', 'SAMEA7697587', 'SAMEA10130100', 'SAMEA13389734', 'SAMEA13389735', 'SAMEA13901728', 'SAMEA13901729', 'SAMEA13929801', 'SAMEA7697622', 'SAMEA10130016', 'SAMEA13389342', 'SAMEA13389343', 'SAMEA13901569', 'SAMEA13901570', 'SAMEA13929771', 'SAMEA7571800', 'SAMEA10130040', 'SAMEA13389503', 'SAMEA13389504', 'SAMEA13901620', 'SAMEA13901621', 'SAMEA13929786', 'SAMEA7571816', 'SAMEA10129979', 'SAMEA13389081', 'SAMEA13389082', 'SAMEA13901489', 'SAMEA13901490', 'SAMEA7571769', 'SAMEA10130002', 'SAMEA13389243', 'SAMEA13389244', 'SAMEA13901540', 'SAMEA13901541', 'SAMEA7697582', 'SAMEA10129985', 'SAMEA13389133', 'SAMEA13389134', 'SAMEA13901505', 'SAMEA13901506', 'SAMEA13929764', 'SAMEA7571775', 'SAMEA10455476', 'SAMEA13389110', 'SAMEA10130031', 'SAMEA13389443', 'SAMEA13389444', 'SAMEA13901602', 'SAMEA13901603', 'SAMEA7571811', 'SAMEA10130023', 'SAMEA13389395', 'SAMEA13389396', 'SAMEA13901586', 'SAMEA13901587', 'SAMEA13929777', 'SAMEA7571806', 'SAMEA10130090', 'SAMEA13389687', 'SAMEA13389688', 'SAMEA13901707', 'SAMEA7571844', 'SAMEA10130119', 'SAMEA13389832', 'SAMEA13389833', 'SAMEA13901773', 'SAMEA13929818', 'SAMEA7697633', 'SAMEA10129996', 'SAMEA13389204', 'SAMEA13389205', 'SAMEA13901529', 'SAMEA13901530', 'SAMEA7697580', 'SAMEA10130088', 'SAMEA13389677', 'SAMEA13389678', 'SAMEA13901704', 'SAMEA13929799', 'SAMEA7571843' mae #> A MultiAssayExperiment object of 8 listed #>  experiments with user-defined names and respective classes. #>  Containing an ExperimentList class object of length 8: #>  [1] BIOGENIC AMINES: TreeSummarizedExperiment with 7 rows and 35 columns #>  [2] FATTY ACIDS: TreeSummarizedExperiment with 19 rows and 57 columns #>  [3] HISTOLOGY: TreeSummarizedExperiment with 20 rows and 57 columns #>  [4] INFLAMMATORY MARKERS: TreeSummarizedExperiment with 14 rows and 58 columns #>  [5] metagenomic_assembly: TreeSummarizedExperiment with 0 rows and 53 columns #>  [6] host_genomic: TreeSummarizedExperiment with 0 rows and 53 columns #>  [7] transcriptomic: TreeSummarizedExperiment with 0 rows and 44 columns #>  [8] metatranscriptomic: TreeSummarizedExperiment with 0 rows and 16 columns #> Functionality: #>  experiments() - obtain the ExperimentList instance #>  colData() - the primary/phenotype DataFrame #>  sampleMap() - the sample coordination DataFrame #>  `$`, `[`, `[[` - extract colData columns, subset, or experiment #>  *Format() - convert into a long or wide DataFrame #>  assays() - convert ExperimentList to a SimpleList of matrices #>  exportClass() - save data to flat files mae[[1]] #> class: TreeSummarizedExperiment  #> dim: 7 35  #> metadata(0): #> assays(1): counts #> rownames(7): Cadaverin Gesamtamine (Total biogenic amines) ... Spermin #>   Tyramin #> rowData names(4): marker.name marker.type marker.canonical_url units #> colnames(35): SAMEA112906114 SAMEA112906592 ... SAMEA112906002 #>   SAMEA112906785 #> colData names(13): accession sample_type ... Project Sample code #> reducedDimNames(0): #> mainExpName: NULL #> altExpNames(0): #> rowLinks: NULL #> rowTree: NULL #> colLinks: NULL #> colTree: NULL"},{"path":"/articles/HoloFoodR.html","id":"incorporate-with-mgnify-data","dir":"Articles","previous_headings":"Functionalities","what":"Incorporate with MGnify data","title":"HoloFoodR: interface to HoloFoodR database","text":"MGnifyR package can utilized fetch metagenomics data MGnify database. MGnifyR package, can use MGnifyR::searchAnalysis() function search analyses based sample IDs . can fetch data based accession IDs. MGnifyR::getResult() returns MAE object just like HoloFoodR. However, metagenomic data points individual analyses instead samples. can harmonize data replacing analysis IDs sample IDs, can combine data single MAE. Now, MAE object linking samples various omics together, compatibility ensured single omics datasets (Tree)SummarizedExperiment format. compatibility allows us harness cutting-edge downstream analytics tools like miaverse framework support data containers seamlessly.","code":"library(MGnifyR)  mg <- MgnifyClient(useCache = TRUE)  # Get those samples that are metagenomic samples metagenomic_samples <- samples[     samples[[\"sample_type\"]] == \"metagenomic_assembly\", ]  # Get analysis IDs based on sample IDs analysis_ids <- searchAnalysis(     mg, type = \"samples\", metagenomic_samples[[\"accession\"]]) head(analysis_ids) #>  SAMEA10130025   SAMEA7697591  SAMEA10130091   SAMEA7571845  SAMEA10158030  #> \"MGYA00606535\" \"MGYA00616692\" \"MGYA00606528\" \"MGYA00616689\" \"MGYA00606518\"  #>   SAMEA7697592  #> \"MGYA00615947\" mae_metagenomic <- MGnifyR::getResult(mg, analysis_ids) mae_metagenomic #> A MultiAssayExperiment object of 6 listed #>  experiments with user-defined names and respective classes. #>  Containing an ExperimentList class object of length 6: #>  [1] microbiota: TreeSummarizedExperiment with 675 rows and 52 columns #>  [2] go-slim: TreeSummarizedExperiment with 116 rows and 52 columns #>  [3] go-terms: TreeSummarizedExperiment with 3264 rows and 52 columns #>  [4] interpro-identifiers: TreeSummarizedExperiment with 19681 rows and 52 columns #>  [5] taxonomy: TreeSummarizedExperiment with 1438 rows and 52 columns #>  [6] taxonomy-lsu: TreeSummarizedExperiment with 1856 rows and 52 columns #> Functionality: #>  experiments() - obtain the ExperimentList instance #>  colData() - the primary/phenotype DataFrame #>  sampleMap() - the sample coordination DataFrame #>  `$`, `[`, `[[` - extract colData columns, subset, or experiment #>  *Format() - convert into a long or wide DataFrame #>  assays() - convert ExperimentList to a SimpleList of matrices #>  exportClass() - save data to flat files # Get experiments from metagenomic data exps <- experiments(mae_metagenomic) # Convert analysis names to sample names exps <- lapply(exps, function(x){     # Get corresponding sample ID     sample_id <- names(analysis_ids)[ match(colnames(x), analysis_ids) ]     # Replace analysis ID with sample ID     colnames(x) <- sample_id     return(x) })  # Add to original MultiAssayExperiment mae <- c(experiments(mae), exps) mae"},{"path":"/articles/HoloFoodR.html","id":"extra-get-data-from-metabolights-database","dir":"Articles","previous_headings":"Functionalities","what":"Extra: Get data from MetaboLights database","title":"HoloFoodR: interface to HoloFoodR database","text":"HoloFood database exclusively contains targeted metabolomic data. However, provides URL addresses linking MetaboLights database, untargeted metabolomics data can accessed. retrieve data, can utilize getMetaboLights() function retrieve information available data. Moreover, returns processed metabolomic data (processed data, can also use getReturn(x, get.metabolomic=TRUE)). , retrieve processed (mapped) metabolomic data associated HoloFood. result list includes three data.frames: spectra data, can either fetch files using getMetaboLightsFile(), follow vignette guidance loading data directly object, tailored metabolomics spectra data.","code":"# Get untargeted metabolomic samples samples <- doQuery(\"samples\", sample_type = \"metabolomic\") # Get the data metabolomic <- getMetaboLights(samples[[\"metabolomics_url\"]])  # Show names of data.frames names(metabolomic) - study metadata - assay metadata - assay that includes abundance table and feature metadata"},{"path":"/articles/HoloFoodR.html","id":"session-info","dir":"Articles","previous_headings":"","what":"Session info","title":"HoloFoodR: interface to HoloFoodR database","text":"","code":"sessionInfo() #> R version 4.4.1 (2024-06-14) #> Platform: x86_64-pc-linux-gnu #> Running under: Ubuntu 22.04.5 LTS #>  #> Matrix products: default #> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  #> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 #>  #> locale: #>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               #>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     #>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    #>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  #>  [9] LC_ADDRESS=C               LC_TELEPHONE=C             #> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        #>  #> time zone: UTC #> tzcode source: system (glibc) #>  #> attached base packages: #> [1] stats4    stats     graphics  grDevices utils     datasets  methods   #> [8] base      #>  #> other attached packages: #>  [1] HoloFoodR_0.99.11               TreeSummarizedExperiment_2.13.0 #>  [3] Biostrings_2.73.2               XVector_0.45.0                  #>  [5] SingleCellExperiment_1.27.2     MultiAssayExperiment_1.31.5     #>  [7] SummarizedExperiment_1.35.4     Biobase_2.65.1                  #>  [9] GenomicRanges_1.57.2            GenomeInfoDb_1.41.2             #> [11] IRanges_2.39.2                  S4Vectors_0.43.2                #> [13] BiocGenerics_0.51.3             MatrixGenerics_1.17.0           #> [15] matrixStats_1.4.1               knitr_1.48                      #> [17] BiocStyle_2.33.1                #>  #> loaded via a namespace (and not attached): #>  [1] httr2_1.0.5             xfun_0.48               bslib_0.8.0             #>  [4] htmlwidgets_1.6.4       lattice_0.22-6          yulab.utils_0.1.7       #>  [7] vctrs_0.6.5             tools_4.4.1             generics_0.1.3          #> [10] curl_5.2.3              parallel_4.4.1          tibble_3.2.1            #> [13] fansi_1.0.6             BiocBaseUtils_1.7.3     pkgconfig_2.0.3         #> [16] Matrix_1.7-0            desc_1.4.3              lifecycle_1.0.4         #> [19] GenomeInfoDbData_1.2.13 compiler_4.4.1          treeio_1.29.1           #> [22] textshaping_0.4.0       codetools_0.2-20        htmltools_0.5.8.1       #> [25] sass_0.4.9              lazyeval_0.2.2          yaml_2.3.10             #> [28] tidyr_1.3.1             pkgdown_2.1.1           pillar_1.9.0            #> [31] crayon_1.5.3            jquerylib_0.1.4         BiocParallel_1.39.0     #> [34] DelayedArray_0.31.14    cachem_1.1.0            abind_1.4-8             #> [37] nlme_3.1-166            tidyselect_1.2.1        digest_0.6.37           #> [40] purrr_1.0.2             dplyr_1.1.4             bookdown_0.41           #> [43] fastmap_1.2.0           grid_4.4.1              cli_3.6.3               #> [46] SparseArray_1.5.45      magrittr_2.0.3          S4Arrays_1.5.11         #> [49] utf8_1.2.4              ape_5.8                 rappdirs_0.3.3          #> [52] UCSC.utils_1.1.0        rmarkdown_2.28          httr_1.4.7              #> [55] ragg_1.3.3              evaluate_1.0.1          rlang_1.1.4             #> [58] Rcpp_1.0.13             tidytree_0.4.6          glue_1.8.0              #> [61] BiocManager_1.30.25     jsonlite_1.8.9          R6_2.5.1                #> [64] systemfonts_1.1.0       fs_1.6.4                zlibbioc_1.51.2"},{"path":"/articles/case_study.html","id":"introduction","dir":"Articles","previous_headings":"","what":"Introduction","title":"HoloFoodR R package — case study","text":"HoloFood database large collection holo-omic multi-omic data two animal systems, chicken salmon. created HoloFood consortium. One objectives study interactions animal systems, microbiomes, feed additives optimize diet strategies farm animals. make easily accessible, creators provide Application Programming Interface (API) permits interaction programming languages, R Python. improve accessibility creating HoloFoodR package simplifies API interaction assists translating raw database data R/Bioconductor data containers, connecting vast ecosystem bioinformatics R packages. aim demonstrate HoloFoodR isolation rest ecosystem, showcase possibility data integration databases, MGnify, holds metagenomic data. Furthermore, provide workflow data exploration advanced machine learning multi-omics, offering practical example readers. main study questions :","code":"- Do treatment associate with salmons' gut flora? - Do larger salmons have different gut microbiota compared to smaller ones? - Do gut microbiota have association with muscle tissue fatty acids? # List of packages that we need packages <- c(     \"dplyr\", \"DT\", \"ggpubr\",  \"ggsignif\", \"HoloFoodR\", \"MGnifyR\", \"mia\", \"miaViz\", \"MOFA2\",     \"patchwork\", \"reticulate\", \"scater\" )  # Load all packages into session. Stop if there are packages that were not # successfully loaded pkgs_not_loaded <- !sapply(packages, require, character.only = TRUE) #> Loading required package: dplyr #>  #> Attaching package: 'dplyr' #> The following objects are masked from 'package:stats': #>  #>     filter, lag #> The following objects are masked from 'package:base': #>  #>     intersect, setdiff, setequal, union #> Loading required package: DT #> Loading required package: ggpubr #> Loading required package: ggplot2 #> Loading required package: ggsignif #> Loading required package: HoloFoodR #> Loading required package: MultiAssayExperiment #> Loading required package: SummarizedExperiment #> Loading required package: MatrixGenerics #> Loading required package: matrixStats #>  #> Attaching package: 'matrixStats' #> The following object is masked from 'package:dplyr': #>  #>     count #>  #> Attaching package: 'MatrixGenerics' #> The following objects are masked from 'package:matrixStats': #>  #>     colAlls, colAnyNAs, colAnys, colAvgsPerRowSet, colCollapse, #>     colCounts, colCummaxs, colCummins, colCumprods, colCumsums, #>     colDiffs, colIQRDiffs, colIQRs, colLogSumExps, colMadDiffs, #>     colMads, colMaxs, colMeans2, colMedians, colMins, colOrderStats, #>     colProds, colQuantiles, colRanges, colRanks, colSdDiffs, colSds, #>     colSums2, colTabulates, colVarDiffs, colVars, colWeightedMads, #>     colWeightedMeans, colWeightedMedians, colWeightedSds, #>     colWeightedVars, rowAlls, rowAnyNAs, rowAnys, rowAvgsPerColSet, #>     rowCollapse, rowCounts, rowCummaxs, rowCummins, rowCumprods, #>     rowCumsums, rowDiffs, rowIQRDiffs, rowIQRs, rowLogSumExps, #>     rowMadDiffs, rowMads, rowMaxs, rowMeans2, rowMedians, rowMins, #>     rowOrderStats, rowProds, rowQuantiles, rowRanges, rowRanks, #>     rowSdDiffs, rowSds, rowSums2, rowTabulates, rowVarDiffs, rowVars, #>     rowWeightedMads, rowWeightedMeans, rowWeightedMedians, #>     rowWeightedSds, rowWeightedVars #> Loading required package: GenomicRanges #> Loading required package: stats4 #> Loading required package: BiocGenerics #>  #> Attaching package: 'BiocGenerics' #> The following objects are masked from 'package:dplyr': #>  #>     combine, intersect, setdiff, union #> The following objects are masked from 'package:stats': #>  #>     IQR, mad, sd, var, xtabs #> The following objects are masked from 'package:base': #>  #>     anyDuplicated, aperm, append, as.data.frame, basename, cbind, #>     colnames, dirname, do.call, duplicated, eval, evalq, Filter, Find, #>     get, grep, grepl, intersect, is.unsorted, lapply, Map, mapply, #>     match, mget, order, paste, pmax, pmax.int, pmin, pmin.int, #>     Position, rank, rbind, Reduce, rownames, sapply, saveRDS, setdiff, #>     table, tapply, union, unique, unsplit, which.max, which.min #> Loading required package: S4Vectors #>  #> Attaching package: 'S4Vectors' #> The following objects are masked from 'package:dplyr': #>  #>     first, rename #> The following object is masked from 'package:utils': #>  #>     findMatches #> The following objects are masked from 'package:base': #>  #>     expand.grid, I, unname #> Loading required package: IRanges #>  #> Attaching package: 'IRanges' #> The following objects are masked from 'package:dplyr': #>  #>     collapse, desc, slice #> Loading required package: GenomeInfoDb #> Loading required package: Biobase #> Welcome to Bioconductor #>  #>     Vignettes contain introductory material; view with #>     'browseVignettes()'. To cite Bioconductor, see #>     'citation(\"Biobase\")', and for packages 'citation(\"pkgname\")'. #>  #> Attaching package: 'Biobase' #> The following object is masked from 'package:MatrixGenerics': #>  #>     rowMedians #> The following objects are masked from 'package:matrixStats': #>  #>     anyMissing, rowMedians #> Loading required package: TreeSummarizedExperiment #> Loading required package: SingleCellExperiment #> Loading required package: Biostrings #> Loading required package: XVector #>  #> Attaching package: 'Biostrings' #> The following object is masked from 'package:base': #>  #>     strsplit #> Loading required package: MGnifyR #>  #> Attaching package: 'MGnifyR' #> The following objects are masked from 'package:HoloFoodR': #>  #>     doQuery, getData, getResult #> Loading required package: mia #> This is mia version 1.13.47 #> - Online documentation and vignettes: https://microbiome.github.io/mia/ #> - Online book 'Orchestrating Microbiome Analysis (OMA)': https://microbiome.github.io/OMA/docs/devel/ #>  #> Attaching package: 'mia' #> The following objects are masked from 'package:dplyr': #>  #>     full_join, inner_join, left_join, right_join #> Loading required package: miaViz #> Loading required package: ggraph #>  #> Attaching package: 'miaViz' #> The following object is masked from 'package:mia': #>  #>     plotNMDS #> Loading required package: MOFA2 #>  #> Attaching package: 'MOFA2' #> The following object is masked from 'package:stats': #>  #>     predict #> Loading required package: patchwork #> Loading required package: reticulate #> Loading required package: scater #> Loading required package: scuttle pkgs_not_loaded <- names(pkgs_not_loaded)[pkgs_not_loaded] if (length(pkgs_not_loaded) > 0) {     stop(\"Error in loading the following packages into the session: '\",         paste0(pkgs_not_loaded, collapse = \"', '\"), \"'\") }"},{"path":"/articles/case_study.html","id":"import-data","dir":"Articles","previous_headings":"","what":"Import data","title":"HoloFoodR R package — case study","text":"start workflow data retrieval. use salmon data associated fatty acid metagenomic amplicon example.","code":""},{"path":"/articles/case_study.html","id":"retrieve-holofood-data","dir":"Articles","previous_headings":"Import data","what":"Retrieve HoloFood data","title":"HoloFoodR R package — case study","text":"First , query HoloFood database retrieve salmon accession numbers. Next, can retrieve data associated salmon. data returned list sample accession numbers associated salmons. example, metagenomic amplicon samples, SAMEA112750580 genomic sample, SAMEA9068691. can use accession numbers fetch data associated sample type store experiments MultiAssayExperiment (MAE) object.","code":"# Get salmon samples salmons <- HoloFoodR::doQuery(\"animals\", system = \"salmon\", use.cache = TRUE)  # Get only the data that has both metagenomic amplicon and fatty acid # data salmons <- salmons |>     filter(fatty_acids == TRUE & metagenomic_amplicon == TRUE)  colnames(salmons) # Get salmon data salmon_data <- HoloFoodR::getData(     accession.type = \"animals\",     accession = salmons[[\"accession\"]],     use.cache = TRUE )  # Get salmon samples salmon_samples <- salmon_data[[\"samples\"]]  # Get sample IDs salmon_sample_ids <- unique(salmon_samples[[\"accession\"]])  head(salmon_sample_ids) # Get salmon <- experiments as MAE object mae <- HoloFoodR::getResult(     salmon_sample_ids,     use.cache = TRUE )"},{"path":"/articles/case_study.html","id":"fetch-metagenomic-data-from-mgnify","dir":"Articles","previous_headings":"Import data","what":"Fetch metagenomic data from MGnify","title":"HoloFoodR R package — case study","text":"HoloFood database include data metagenomic data. data can retrieved MGnify portal. purpose, use MGnifyR package, similar fashion HoloFoodR, allows simple interaction MGnify API. salmon_analysis_ids character vector holds associations HoloFood metagenomic amplicon accession numbers (SAMEAxxxxxx) counterparts MGnify database (MGYAxxxxxx). Data fetched MGnify MGnify-specific identifiers. first rename samples HoloFood specific ID add data MultiAssayExperiment combining data. Now retrieved data interested workflow.","code":"# Create MGnify object mg <- MgnifyClient(     useCache = TRUE,     cacheDir = \".MGnifyR_cache\" )  # Select only metagenomic_amplicon sample type metagenomic_salmon_samples <- salmon_samples |>     filter(sample_type == \"metagenomic_amplicon\")  # Search for sample IDs in MGnify database salmon_analysis_ids <- searchAnalysis(     mg,     type = \"samples\",     metagenomic_salmon_samples[[\"accession\"]] ) # Get metagenomic taxonomic data for salmon from MGnify tse <- MGnifyR::getResult(     mg,     accession = salmon_analysis_ids,     get.func = FALSE ) # Add MGnify results to HoloFood data mae <- addMGnify(tse, mae) #> Warning: 'experiments' dropped; see 'drops()' #> harmonizing input: #>   removing 189 sampleMap rows not in names(experiments) #> harmonizing input: #>   removing 189 sampleMap rows not in names(experiments)"},{"path":"/articles/case_study.html","id":"data-preprocess","dir":"Articles","previous_headings":"","what":"Data preprocess","title":"HoloFoodR R package — case study","text":"Data cleaning one time-consuming important steps data analysis. instance, need handle missing data, transform data assays, agglomerate data. next steps, : Filter data Ensure data correct format analysis Agglomerate data Transform data","code":""},{"path":"/articles/case_study.html","id":"wrangle-the-data","dir":"Articles","previous_headings":"Data preprocess","what":"Wrangle the data","title":"HoloFoodR R package — case study","text":"see upset plot summarizes available experiments samples overlap .  demonstration purposes, focus investigating fatty acids metagenomic data (fatty_acids_mg metagenomic_amplicon) within trial performed HoloFood consortium. trial health effects fermented seaweed added diet salmons. values fatty acids detection thresholds. assume zeroes. Moreover, data includes feature just states fatty acids collected. remove feature assay contains numeric values. Moreover, wrangle sample metadata includes necessary information. want analyze effect treatment, must analyze salmons day 60 salmons sampled experimental period. contrast, day-0 salmons sampled immediately beginning trial. select samples drawn day 60. Now data includes samples taken day 60. animal microbiota data fatty acid data.","code":"upset_plot <- upsetSamples(mae) upset_plot # Harmonize experiment names names(mae) <- names(mae) |>     tolower() |>     gsub(pattern = \" \", replacement = \"_\")  # Fetch only experiments that we need mae <- mae[, , c(\"fatty_acids_mg\", \"metagenomic\")] #> Warning: 'experiments' dropped; see 'drops()' #> harmonizing input: #>   removing 934 sampleMap rows not in names(experiments) names(mae) <- c(\"fatty_acids\", \"metagenomic\") # Filter MAE object to include only Trial A mae <- mae[, colData(mae)[[\"Trial code\"]] == \"SA\", ] # From metabolomic data, remove organ-fatty acids row because it only contains a # string value \"muscle\" which denotes where the sample was drawn. tse <- mae[[1]] tse <- tse[!(rowData(tse)[[\"marker.name\"]] %in% c(\"Organ-fatty acids\")), ] mae[[1]] <- tse  # Transform matrix to numeric. Some values are \"< 0.01\" # If a number is < 0.01, assume it to be 0 assay <- assay(mae[[1]], \"counts\") assay[assay == \"<0.01\"] <- 0 assay <- apply(assay, c(1, 2), function(x) as.numeric(gsub(\",\", \".\", x)))  # Reassign assay back to MAE assay(mae[[1]], \"counts\") <- assay # Add treatment groups colData(mae)[[\"treatment_group\"]] <- ifelse(   colData(mae)[[\"Treatment concentration\"]]>0, \"treatment\", \"control\")  # Add salmon host mass to MAE colData colData(mae[[2]])[[\"host.gutted.mass\"]] <- as.numeric(     colData(mae[[2]])[[\"host.gutted.mass\"]]) colData(mae[[2]])[[\"mass_scaled\"]] <- scale(   colData(mae[[2]])[[\"host.gutted.mass\"]]) merged_data <- merge(     x = colData(mae),     y = colData(mae[[2]])[, c(\"animal\", \"host.gutted.mass\", \"mass_scaled\")],     all.x = TRUE, by = \"animal\" ) rownames(merged_data) <- merged_data[[\"animal\"]] # Order and add back colData(mae) <- merged_data[match(     colData(mae)[[\"animal\"]], merged_data[[\"animal\"]]), ]  # Add animal metadata to separate experiments mae[[1]] <- getWithColData(mae, 1) #> Warning: Ignoring redundant column names in 'colData(x)': canonical_url, #> marker.type mae[[2]] <- getWithColData(mae, 2) #> Warning: Ignoring redundant column names in 'colData(x)': canonical_url, marker.type, #>   mass_scaled # Subset by taking only samples from day 60 samples <- colnames(mae[[2]])[mae[[2]]$trial.timepoint == 60] mae[[2]] <- mae[[2]][, samples] #> harmonizing input: #>   removing 15 sampleMap rows with 'colname' not in colnames of experiments # Make sure that we have same set of samples from each experiment mae <- intersectColumns(mae)"},{"path":"/articles/case_study.html","id":"filtering-and-agglomeration","dir":"Articles","previous_headings":"Data preprocess","what":"Filtering and agglomeration","title":"HoloFoodR R package — case study","text":"Next, can agglomerate features prevalence reduce number low-abundant taxa contaminants. First, visualize prevalence distribution taxa histogram decide prevalence threshold use. use 0.1% detection level filter extremely low-abundant genera.  can also look raw prevalence numbers Mycoplasma present samples, surprising genus found one common salmon intestine (see Zarkasi et al. (2014)). use thresholds 20% 0.1% prevalence detection, respectively. Moreover, agglomerate data include 19 abundant genera. Due limited number samples, also filter fatty acid data include fatty acids show variation within dataset. rationale fatty acid vary, exhibit differences groups. can find good threshold cut-histogram standard deviations fatty acid abundances.  apply filtering threshold 0.5 exclude fatty acids exhibit sufficient variation dataset. Since standard deviations 1, reasonable number want exclude many fatty acids. detailed analysis, pick certain fatty acids well-established biological relevance. include 22:6n-3, 20:5n-3, 18:3n-3, 20:4n-6, 18:2n-6, 18:1n-9, 16:0, 18:0.","code":"# Add relative transforrmed data mae[[2]] <- transformAssay(mae[[2]], method = \"relabundance\")  prevalence <- getPrevalence(     mae[[2]],     rank = \"Genus\",     assay.type = \"relabundance\",     na.rm = TRUE,     sort = TRUE,     detection = 0.1 / 100 )  # Exclude microbes with 0 prevalence prevalence <- prevalence[prevalence != 0]  hist(prevalence, main = \"Prevalence of microbial taxa\",      xlab = \"Prevalence\") # Sort prevalence in decreasing order sort(prevalence, decreasing = TRUE) |> head(10) #>     Mycoplasma Photobacterium     Aliivibrio  Cetobacterium         Vibrio  #>      1.0000000      0.9666667      0.6666667      0.6000000      0.5000000  #> Ammopiptanthus  Mycobacterium       Medicago    Paucibacter   Enterococcus  #>      0.2000000      0.1666667      0.1333333      0.1000000      0.1000000 # Agglomerate by prevalence by genus altExp(mae[[2]], \"prev_genus\") <- agglomerateByPrevalence(     mae[[2]],     assay.type = \"relabundance\",     rank = \"Genus\",     prevalence = 0.2 / 100,     detection = 0.1 / 100 ) tse <- agglomerateByRank(mae[[2]], rank = \"Genus\") core <- getTop(tse, top = 19) # Get vector that tells whether taxa belongs to core taxa or not core_names <- rownames(tse) core_names[ !core_names %in% core ] <- \"other\" # Agglomerate based on core taxa rowData(tse)[[\"core\"]] <- core_names altExp(mae[[2]], \"core_genus\") <- agglomerateByVariable(   tse, by = \"rows\", group = \"core\") rowData(mae[[1]])[[\"sd\"]] <- rowSds(assay(mae[[1]], \"counts\"), na.rm = TRUE) hist(rowData(mae[[1]])[[\"sd\"]],     main = \"Standard deviation distribution of fatty acid abundances\",     xlab = \"Standard deviation\",     breaks = 20) mae[[1]] <- mae[[1]][ rowData(mae[[1]])[[\"sd\"]] > 0.5, ] relevant_fatty_acids <- c(   \"Docosahexaenoic acid 22:6n-3 (DHA)\",   \"Eicosapentaenoic acid 20:5n-3 (EPA)\",   \"Alpha-Linolenic acid 18:3n-3\",   \"Arachidonic acid 20:4n-6 (ARA)\",   \"Linoleic acid 18:2n-6\",   \"Oleic acid 18:1n-9\",   \"Palmitic acid 16:0\",   \"Stearic acid 18:0\" ) altExp(mae[[1]], \"relevant\") <- mae[[1]][     rownames(mae[[1]]) %in% relevant_fatty_acids, ]"},{"path":"/articles/case_study.html","id":"transformation","dir":"Articles","previous_headings":"Data preprocess","what":"Transformation","title":"HoloFoodR R package — case study","text":"transform metagenomic counts relative transformation centered log-ratio method tackle compositional data (see Quinn et al. (2019)). Fatty acid data already compositional measured concentration (mg/g). apply log10 transformation address skewness data. Finally, data standardized ensure features comparable scale.","code":"# Transform microbiome with centered log-ratio method mae[[2]] <- transformAssay(     mae[[2]],     assay.type = \"counts\",     method = \"relabundance\",     MARGIN = \"cols\" ) mae[[2]] <- transformAssay(     mae[[2]],     assay.type = \"counts\",     method = \"clr\",     pseudocount = TRUE,     MARGIN = \"cols\" ) #> A pseudocount of 0.5 was applied. #> A pseudocount of 0.5 was applied. #> A pseudocount of 0.5 was applied. mae[[1]] <- transformAssay(     mae[[1]],     assay.type = \"counts\",     method = \"log10\",     MARGIN = \"cols\" ) mae[[1]] <- transformAssay(     mae[[1]],     assay.type = \"log10\",     method = \"standardize\",     MARGIN = \"rows\" )"},{"path":[]},{"path":"/articles/case_study.html","id":"does-treatment-affect-fatty-acids","dir":"Articles","previous_headings":"Fatty acids, mass and treatment","what":"Does treatment affect fatty acids?","title":"HoloFoodR R package — case study","text":"aim investigate whether treatment associated changes fatty acid concentrations. begin, assess correlation treatment levels concentrations specific fatty acids observed, treatment appears significant effect fatty acid concentrations.","code":"# Get fatty acids that we are going to test tse <- altExp(mae[[1]], \"relevant\")  # For each fatty acid, test if the groups are different res <- lapply(rownames(tse), function(feat){     res <- pairwise.wilcox.test(       assay(tse, \"counts\")[feat, ], tse[[\"treatment_group\"]],       p.adjust.method = \"none\")     return(res) }) #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties #> Warning in wilcox.test.default(xi, xj, paired = paired, ...): cannot compute #> exact p-value with ties # Combine results and adjust p -values res <- do.call(rbind, res) |> as.data.frame() res[[\"p.value\"]] <- p.adjust(res[[\"p.value\"]], method = \"fdr\") res |> datatable()"},{"path":"/articles/case_study.html","id":"is-there-association-between-salmon-size-and-fatty-acids","dir":"Articles","previous_headings":"Fatty acids, mass and treatment","what":"Is there association between salmon size and fatty acids?","title":"HoloFoodR R package — case study","text":"Additionally, worth exploring whether size salmon associated variations fatty acid levels. appears notable effect: larger salmon seem slightly higher concentrations fatty acids. trend can illustrated plots.  result initial analysis fatty acids, discovered size salmon appears association fatty acid concentrations, although association statistically significant.","code":"# Calculate correlation res <- getCrossAssociation(     mae, experiment1 = 1, experiment2 = 1, altexp1 = \"relevant\",     assay.type1 = \"standardize\", col.var2 = \"mass_scaled\",     test.signif = TRUE, method = \"pearson\" ) #> Calculating correlations... #> altexp1: relevant, altexp2: -, assay.type1: standardize, col.var1: -, assay.type2: -, col.var2: mass_scaled #> by: 1, function: stats::cor.test, method: pearson, test.signif: TRUE, p.adj.method: fdr, paired: FALSE, show.warnings: TRUE res #>                                  Var1        Var2       cor        pval #> 1  Docosahexaenoic acid 22:6n-3 (DHA) mass_scaled 0.4694512 0.008864359 #> 2 Eicosapentaenoic acid 20:5n-3 (EPA) mass_scaled 0.3666143 0.046294243 #> 3               Linoleic acid 18:2n-6 mass_scaled 0.3521072 0.056359869 #> 4                  Oleic acid 18:1n-9 mass_scaled 0.3585233 0.051714669 #> 5                  Palmitic acid 16:0 mass_scaled 0.4616248 0.010233011 #>        p_adj #> 1 0.02558253 #> 2 0.05635987 #> 3 0.05635987 #> 4 0.05635987 #> 5 0.02558253 p <- plotExpression(     altExp(mae[[1]], \"relevant\"),     features = rownames(altExp(mae[[1]], \"relevant\")),     x = \"mass_scaled\",     assay.type = \"standardize\",     scales = \"free\"     ) +   geom_smooth(method = \"lm\") p #> `geom_smooth()` using formula = 'y ~ x'"},{"path":[]},{"path":"/articles/case_study.html","id":"microbial-composition","dir":"Articles","previous_headings":"Microbiota, mass and treatment","what":"Microbial composition","title":"HoloFoodR R package — case study","text":"first step analysing microbiota data, summarize microbial composition relative abundance barplot.  Salmon gut seems dominated either genus Mycoplasma Photobacterium.","code":"p <- plotAbundance(     altExp(mae[[2]], \"core_genus\"),     assay.type = \"relabundance\",     col.var = \"treatment_group\",     facet.cols = TRUE, scales = \"free_x\"     ) p #> Warning: Removed 200 rows containing missing values or values outside the scale range #> (`geom_bar()`)."},{"path":"/articles/case_study.html","id":"is-alpha-diversity-associated-with-treatment-or-mass-of-salmon","dir":"Articles","previous_headings":"Microbiota, mass and treatment","what":"Is alpha diversity associated with treatment or mass of salmon?","title":"HoloFoodR R package — case study","text":"Next, let’s calculate Shannon alpha diversity index test whether diversity associated size treatment.  , observe potential effect, time treatment microbial diversity, although statistically significant. Additionally, association diversity mass. Let’s analyze whether can find similar effect beta diversity. perform Principal Coordinate Analysis (PCoA) Bray-Curtis dissimilarity.  seems handful samples differ others. However, effect seems associated treatment group.","code":"# Calculate alpha diversity mae[[2]] <- addAlpha(mae[[2]]) #> Warning: 'faith' index can be calculated only for TreeSE with rowTree(x) #> populated or with 'tree' provided separately.  # Plot with metadata p1 <- plotColData(mae[[2]], x = \"mass_scaled\", y = \"shannon_diversity\") +   geom_smooth(method = \"lm\") +   ggtitle(\"Shannon diversity association\\nwith body mass\") +   xlab(\"Scaled body mass\") + ylab(\"Shannon diversity index\") p2 <- plotColData(mae[[2]], x = \"treatment_group\", y = \"shannon_diversity\") +     geom_signif(comparisons = list(c(\"control\", \"treatment\")), ) +   ggtitle(\"Shannon diversity association\\nwith treatment\") +   xlab(\"Treatment group\") + ylab(\"Shannon diversity index\")  p1 + p2 #> `geom_smooth()` using formula = 'y ~ x' # Run PCoA mae[[2]] <- runMDS(     mae[[2]],     FUN = getDissimilarity,     method = \"bray\",     assay.type = \"relabundance\" )  # Display dissimilarity on a plot p <- plotReducedDim(mae[[2]], \"MDS\", colour_by = \"treatment_group\") +   ggtitle(\"PCoA plot of Bray-Curtis dissimilarity by treatment group\") p"},{"path":"/articles/case_study.html","id":"multi-omics-integration","dir":"Articles","previous_headings":"","what":"Multi-omics integration","title":"HoloFoodR R package — case study","text":"Multi-omic factor analysis (MOFA) (see Argelaguet et al. (2020)). allows us discover latent factors underlie biological differences taking consideration 2 omic assays. cite original authors, “MOFA can viewed statistically rigorous generalization (sparse) principal component analysis (PCA) multi-omics data”. applying MOFA analysis, goal determine whether metagenomics fatty acids exhibit shared variability, ultimately assessing whether microbial community associated fatty acid composition. Next, plot variances explained factor.  Factor 1 captures mostly variance metagenomics data. Factor 2 captures shared variability metagenomics fatty acid data. first factor captures highest amount variability metagenomics data, almost much variability remains captured second factor. exploring shared variability, first examine taxa’s variability captured Factor 1.  plot , can see first factor captures mostly variability Mycoplasma.  microbes, especially Photobacterium, positive association Factor 2. Many fatty acids also show significant weights, Linoleic acid 18:2n-6 sum omega-6 fatty acids positive weights, among others. indicates abundance Photobacterium increases, corresponding increase levels aforementioned fatty acids.","code":"mae_temp <- mae mae_temp[[2]] <- altExp(mae_temp[[2]], \"prev_genus\")  # Extract only transformed metagenomic assays for MOFA analysis assays(mae_temp[[1]]) <- assays(mae_temp[[1]])[     names(assays(mae_temp[[1]])) %in% c(\"standardize\") ] assays(mae_temp[[2]]) <- assays(mae_temp[[2]])[     names(assays(mae_temp[[2]])) %in% c(\"counts\") ]  # Transform MAE object to MOFA model model <- create_mofa_from_MultiAssayExperiment(mae_temp)  # Set model's options model_opts <- get_default_model_options(model) model_opts$num_factors <- 5 model_opts$likelihoods[[2]] <- \"poisson\" train_opts <- get_default_training_options(model) train_opts$maxiter <- 20000  # Change convergence mode to slightly improve accuracy train_opts$convergence_mode <- \"slow\"  # Prepare MOFA model model <- prepare_mofa(     object = model,     model_options = model_opts,     training_options = train_opts ) #> Checking data options... #> No data options specified, using default... #> Checking training options... #> Checking model options...  # Train model model <- run_mofa(model, use_basilisk = TRUE) #> Warning in run_mofa(model, use_basilisk = TRUE): No output filename provided. Using /tmp/RtmpMdYbfu/mofa_20241024-132828.hdf5 to store the trained model. #> Connecting to the mofapy2 package using basilisk.  #>     Set 'use_basilisk' to FALSE if you prefer to manually set the python binary using 'reticulate'. p <- plot_variance_explained(model) +     labs(title = \"Captured variance of fatty acid and metagenomic\\n\",         \"datasets across 5 factors\") +     theme(plot.title.position = \"plot\") p p1 <- plot_top_weights(model, view = 1, factors = 1, nfeatures = 25) +   labs(title = \"Fatty acids\") p2 <- plot_top_weights(model, view = 2, factors = 1, nfeatures = 25) +   labs(title = \"Microbiota\")  p1 + p2 p1 <- plot_top_weights(model, view = 1, factors = 2, nfeatures = 25) +   labs(title = \"Fatty acids\") p2 <- plot_top_weights(model, view = 2, factors = 2, nfeatures = 25) +   labs(title = \"Microbiota\")  p1 + p2"},{"path":"/articles/case_study.html","id":"conclusions","dir":"Articles","previous_headings":"","what":"Conclusions","title":"HoloFoodR R package — case study","text":"present case study demonstrated easy fast download large dataset transform data MultiAssayExperiment, turn gives researchers access extensive plethora downstream tools, mia MOFA2 can used pre-process visualize multi-omics data.","code":"sessionInfo() #> R version 4.4.1 (2024-06-14) #> Platform: x86_64-pc-linux-gnu #> Running under: Ubuntu 22.04.5 LTS #>  #> Matrix products: default #> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3  #> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so;  LAPACK version 3.10.0 #>  #> locale: #>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               #>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8     #>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    #>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                  #>  [9] LC_ADDRESS=C               LC_TELEPHONE=C             #> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C        #>  #> time zone: UTC #> tzcode source: system (glibc) #>  #> attached base packages: #> [1] stats4    stats     graphics  grDevices utils     datasets  methods   #> [8] base      #>  #> other attached packages: #>  [1] scater_1.33.4                   scuttle_1.15.4                  #>  [3] reticulate_1.39.0               patchwork_1.3.0                 #>  [5] MOFA2_1.15.0                    miaViz_1.13.15                  #>  [7] ggraph_2.2.1                    mia_1.13.47                     #>  [9] MGnifyR_1.1.0                   HoloFoodR_0.99.11               #> [11] TreeSummarizedExperiment_2.13.0 Biostrings_2.73.2               #> [13] XVector_0.45.0                  SingleCellExperiment_1.27.2     #> [15] MultiAssayExperiment_1.31.5     SummarizedExperiment_1.35.4     #> [17] Biobase_2.65.1                  GenomicRanges_1.57.2            #> [19] GenomeInfoDb_1.41.2             IRanges_2.39.2                  #> [21] S4Vectors_0.43.2                BiocGenerics_0.51.3             #> [23] MatrixGenerics_1.17.0           matrixStats_1.4.1               #> [25] ggsignif_0.6.4                  ggpubr_0.6.0                    #> [27] ggplot2_3.5.1                   DT_0.33                         #> [29] dplyr_1.1.4                     knitr_1.48                      #> [31] BiocStyle_2.33.1                #>  #> loaded via a namespace (and not attached): #>   [1] fs_1.6.4                    tidyjson_0.3.2              #>   [3] DirichletMultinomial_1.47.2 httr_1.4.7                  #>   [5] RColorBrewer_1.1-3          tools_4.4.1                 #>   [7] backports_1.5.0             utf8_1.2.4                  #>   [9] R6_2.5.1                    vegan_2.6-8                 #>  [11] HDF5Array_1.33.8            lazyeval_0.2.2              #>  [13] uwot_0.2.2                  mgcv_1.9-1                  #>  [15] rhdf5filters_1.17.0         permute_0.9-7               #>  [17] withr_3.0.1                 gridExtra_2.3               #>  [19] cli_3.6.3                   textshaping_0.4.0           #>  [21] sandwich_3.1-1              labeling_0.4.3              #>  [23] slam_0.1-54                 sass_0.4.9                  #>  [25] mvtnorm_1.3-1               pkgdown_2.1.1               #>  [27] systemfonts_1.1.0           yulab.utils_0.1.7           #>  [29] foreign_0.8-87              decontam_1.25.0             #>  [31] rstudioapi_0.17.1           generics_0.1.3              #>  [33] gridGraphics_0.5-1          crosstalk_1.2.1             #>  [35] car_3.1-3                   rbiom_1.0.3                 #>  [37] Matrix_1.7-0                ggbeeswarm_0.7.2            #>  [39] fansi_1.0.6                 DECIPHER_3.1.5              #>  [41] abind_1.4-8                 lifecycle_1.0.4             #>  [43] yaml_2.3.10                 carData_3.0-5               #>  [45] rhdf5_2.49.0                SparseArray_1.5.45          #>  [47] Rtsne_0.17                  grid_4.4.1                  #>  [49] crayon_1.5.3                dir.expiry_1.13.1           #>  [51] lattice_0.22-6              beachmat_2.21.8             #>  [53] cowplot_1.1.3               pillar_1.9.0                #>  [55] boot_1.3-31                 lpSolve_5.6.21              #>  [57] codetools_0.2-20            glue_1.8.0                  #>  [59] ggfun_0.1.7                 data.table_1.16.2           #>  [61] vctrs_0.6.5                 png_0.1-8                   #>  [63] treeio_1.29.1               urltools_1.7.3              #>  [65] gtable_0.3.5                assertthat_0.2.1            #>  [67] cachem_1.1.0                xfun_0.48                   #>  [69] S4Arrays_1.5.11             tidygraph_1.3.1             #>  [71] pheatmap_1.0.12             bluster_1.15.1              #>  [73] nlme_3.1-166                ggtree_3.13.1               #>  [75] filelock_1.0.3              UpSetR_1.4.0                #>  [77] SnowballC_0.7.1             bslib_0.8.0                 #>  [79] irlba_2.3.5.1               vipor_0.4.7                 #>  [81] rpart_4.1.23                colorspace_2.1-1            #>  [83] DBI_1.2.3                   Hmisc_5.1-3                 #>  [85] nnet_7.3-19                 tidyselect_1.2.1            #>  [87] compiler_4.4.1              httr2_1.0.5                 #>  [89] htmlTable_2.4.3             BiocNeighbors_1.99.3        #>  [91] basilisk.utils_1.17.3       desc_1.4.3                  #>  [93] DelayedArray_0.31.14        bookdown_0.41               #>  [95] triebeard_0.4.1             checkmate_2.3.2             #>  [97] scales_1.3.0                rappdirs_0.3.3              #>  [99] stringr_1.5.1               digest_0.6.37               #> [101] minqa_1.2.8                 rmarkdown_2.28              #> [103] basilisk_1.17.2             htmltools_0.5.8.1           #> [105] pkgconfig_2.0.3             base64enc_0.1-3             #> [107] lme4_1.1-35.5               sparseMatrixStats_1.17.2    #> [109] highr_0.11                  fastmap_1.2.0               #> [111] rlang_1.1.4                 htmlwidgets_1.6.4           #> [113] UCSC.utils_1.1.0            DelayedMatrixStats_1.27.3   #> [115] farver_2.1.2                jquerylib_0.1.4             #> [117] zoo_1.8-12                  jsonlite_1.8.9              #> [119] BiocParallel_1.39.0         tokenizers_0.3.0            #> [121] BiocSingular_1.21.4         magrittr_2.0.3              #> [123] Formula_1.2-5               GenomeInfoDbData_1.2.13     #> [125] ggplotify_0.1.2             Rhdf5lib_1.27.0             #> [127] munsell_0.5.1               Rcpp_1.0.13                 #> [129] ape_5.8                     ggnewscale_0.5.0            #> [131] viridis_0.6.5               stringi_1.8.4               #> [133] zlibbioc_1.51.2             MASS_7.3-61                 #> [135] plyr_1.8.9                  mediation_4.5.0             #> [137] parallel_4.4.1              ggrepel_0.9.6               #> [139] forcats_1.0.0               graphlayouts_1.2.0          #> [141] splines_4.4.1               igraph_2.1.1                #> [143] reshape2_1.4.4              ScaledMatrix_1.13.0         #> [145] evaluate_1.0.1              tidytext_0.4.2              #> [147] RcppParallel_5.1.9          BiocManager_1.30.25         #> [149] nloptr_2.1.1                tweenr_2.0.3                #> [151] tidyr_1.3.1                 purrr_1.0.2                 #> [153] polyclip_1.10-7             BiocBaseUtils_1.7.3         #> [155] ggforce_0.4.2               rsvd_1.0.5                  #> [157] broom_1.0.7                 tidytree_0.4.6              #> [159] janeaustenr_1.0.0           rstatix_0.7.2               #> [161] viridisLite_0.4.2           ragg_1.3.3                  #> [163] tibble_3.2.1                aplot_0.2.3                 #> [165] memoise_2.0.1               beeswarm_0.4.0              #> [167] cluster_2.1.6               corrplot_0.95"},{"path":"/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Tuomas Borman. Author, maintainer. Artur Sannikov. Author. Leo Lahti. Author.","code":""},{"path":"/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Borman T, Sannikov , Lahti L (2024). HoloFoodR: R interface EBI HoloFood resource. R package version 0.99.11, https://github.com/EBI-Metagenomics/HoloFoodR.","code":"@Manual{,   title = {HoloFoodR: R interface to EBI HoloFood resource},   author = {Tuomas Borman and Artur Sannikov and Leo Lahti},   year = {2024},   note = {R package version 0.99.11},   url = {https://github.com/EBI-Metagenomics/HoloFoodR}, }"},{"path":"/index.html","id":"holofoodr-","dir":"","previous_headings":"","what":"R interface to EBI HoloFood resource","title":"R interface to EBI HoloFood resource","text":"R package developed streamlining integration analysis EMBL-EBI HoloFood data. utility package simplifies access resource, enabling direct loading data formats tailored multiomics downstream analytics. tool, users can efficiently search retrieve data EBI HoloFood resource. retrieved data available MultiAssayExperiment / TreeSummarizedExperiment format similarly data acquired MGnifyR package MGnify database. compatibility ensures users can seamlessly combine analyze datasets sources, leading valuable insights intricate biological systems. research received funding Horizon 2020 Programme European Union within FindingPheno project grant agreement 952914.","code":""},{"path":"/index.html","id":"findingpheno-","dir":"","previous_headings":"","what":"FindingPheno","title":"R interface to EBI HoloFood resource","text":"FindingPheno, EU-funded project, dedicated developing computational tools methodologies integration analysis multi-omics data. primary objective deepen understanding interactions hosts microbiomes. FindingPheno website","code":""},{"path":"/index.html","id":"holofood-","dir":"","previous_headings":"","what":"HoloFood","title":"R interface to EBI HoloFood resource","text":"HoloFood, project funded EU’s Horizon 2020 programme, employed holistic, “hologenomic”, approach enhance efficiency food production systems. involved exploring biomolecular physiological processes triggered incorporation feed additives novel sustainable feeds farmed animals. HoloFood database, hosted European Bioinformatics Institute (EMBL-EBI), houses data gathered project, encompassing multiple omics, including metabolomics various biomolecular measurements. Notably, include data metagenomic untargeted metabolomic analyses. However, metagenomic data project can accessed MGnify database, untargeted metabolomic data stored MetaboLights database. explore available datasets HoloFood, can utilize API browser. HoloFood website HoloFood publications API browser API documentation","code":""},{"path":"/index.html","id":"mgnify-","dir":"","previous_headings":"","what":"MGnify","title":"R interface to EBI HoloFood resource","text":"EMBL-EBI’s MGnify serves repository microbiome data, offering wide array analyses encompassing metabarcoding, metatranscriptomic, metagenomic datasets diverse environments. platform provides comprehensive taxonomic assignments functional annotations datasets. data can accessed MGnifyR package. MGnify website MGnifyR","code":""},{"path":"/index.html","id":"metabolights-","dir":"","previous_headings":"","what":"MetaboLights","title":"R interface to EBI HoloFood resource","text":"MetaboLights, managed EMBL-EBI, serves repository metabolomic data. can accessed HoloFoodR package. MetaboLights website","code":""},{"path":"/index.html","id":"miaverse-","dir":"","previous_headings":"","what":"miaverse","title":"R interface to EBI HoloFood resource","text":"miaverse R/Bioconductor framework specialized microbiome downstream data analysis, leveraging TreeSummarizedExperiment class. offers comprehensive suite tools microbiome bioinformatics. Additionally, miaverse includes tutorial book Orchestrating Microbiome Analysis (OMA), aims guide users conducting microbiome data analysis sharing best practices microbiome data science. miaverse website Orchestrating Microbiome Analysis (OMA)","code":""},{"path":[]},{"path":"/index.html","id":"bioc-release","dir":"","previous_headings":"Installation","what":"Bioc-release","title":"R interface to EBI HoloFood resource","text":"","code":"if (!requireNamespace(\"BiocManager\", quietly = TRUE))     install.packages(\"BiocManager\")  BiocManager::install(\"HoloFoodR\")"},{"path":"/index.html","id":"bioc-devel","dir":"","previous_headings":"Installation","what":"Bioc-devel","title":"R interface to EBI HoloFood resource","text":"","code":"if (!requireNamespace(\"BiocManager\", quietly = TRUE))     install.packages(\"BiocManager\")  # The following initializes usage of Bioc devel BiocManager::install(version=\"devel\")  BiocManager::install(\"HoloFoodR\")"},{"path":"/index.html","id":"github","dir":"","previous_headings":"Installation","what":"GitHub","title":"R interface to EBI HoloFood resource","text":"","code":"remotes::install_github(\"EBI-Metagenomics/HoloFoodR\")"},{"path":"/index.html","id":"basic-usage","dir":"","previous_headings":"","what":"Basic usage","title":"R interface to EBI HoloFood resource","text":"detailed instructions read associated function help, function reference page vignette (vignette(\"HoloFoodR\"))","code":"library(HoloFoodR)  # Search samples samples <- doQuery(\"samples\")  # Search animals animals <- doQuery(\"animal\")  # Fetch data on certain sample samples <- c(\"ACCESSION_ID\") sample_data <- getData(accession.type = \"samples\", accession = samples)  # Fetch data on genome catalogues genome_catalogues <- getData(type = \"genome-catalogues\")  # Fetch data on genomes in certain genome catalogue catalogues <- c(\"ACCESSION_ID\") genomes <- getData(     type = \"genomes\",     accession.type = \"genome-catalogues\",     accession = catalogues)  # Fetch data on untargeted metabolites metabolites <- getMetaboLights(study_id)  # Fetch data as MultiAssayExperiment samples <- c(\"ACCESSION_ID\") mae <- getResult(accession = samples)"},{"path":"/reference/HoloFoodR.html","id":null,"dir":"Reference","previous_headings":"","what":"HoloFoodR package — HoloFoodR","title":"HoloFoodR package — HoloFoodR","text":"HoloFoodR implements interface EBI HoloFood database. See vignette general introduction package, [HoloFood](https://www.holofood.eu/) general HoloFood information, [API documentation](https://docs.holofooddata.org/api.html) details JSONAPI implementation.","code":""},{"path":[]},{"path":"/reference/HoloFoodR.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"HoloFoodR package — HoloFoodR","text":"Maintainer: Tuomas Borman tuomas.v.borman@utu.fi (ORCID) Authors: Artur Sannikov arsann@utu.fi (ORCID) Leo Lahti leo.lahti@iki.fi (ORCID)","code":""},{"path":"/reference/addMGnify.html","id":null,"dir":"Reference","previous_headings":"","what":"Add results from MGnifyR to HoloFoodR results — addMGnify","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"Add results MGnifyR HoloFoodR results","code":""},{"path":"/reference/addMGnify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"","code":"addMGnify(x, y, ...)  # S4 method for class 'SummarizedExperiment,MultiAssayExperiment' addMGnify(   x,   y,   exp.name1 = \"metagenomic\",   exp.name2 = \"metagenomic_amplicon\",   replace = TRUE,   ... )  # S4 method for class 'SummarizedExperiment,SummarizedExperiment' addMGnify(x, y, ...)  # S4 method for class 'SummarizedExperiment,ANY' addMGnify(x, y, id.col1 = \"sample_biosample\", id.col2 = \"accession\", ...)"},{"path":"/reference/addMGnify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"x SummarizedExperiment. Results MGnifyR::getResult(). y MultiAssayExperiment SummarizedExperiment data.frame-like table. Results HoloFoodR::getResult() sample metadata . ... optional arguments used currently. exp.name1 Character scalar. Specifies name experiment added y. (Default: \"metagenomic\") exp.name2 Character scalar. Specifies name experiment HoloFoodR results. experiment used match IDs MGnify data. (Default: \"metagenomic_amplicon\") replace Logical scalar. Whether replace template experiment. (Default: TRUE) id.col1 Character scalar. Specifies name column colData(x) includes HoloFood identifiers. (Default: \"sample_biosample\") id.col2 Character scalar. Specifies name column colData(y[[exp.name2]]) includes HoloFood identifiers. (Default: \"accession\")","code":""},{"path":"/reference/addMGnify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"MultiAssayExperiment","code":""},{"path":"/reference/addMGnify.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"Metagenomic data found MGnify rather HoloFoodR, two databases use different sample identifiers. However, MGnify's sample metadata includes references identifiers used HoloFood database, making straightforward convert sample IDs alignment HoloFood data. Despite , HoloFood contains additional metadata available MGnify. Moreover, integrating data MultiAssayExperiment maintaining accurate sample system matches can challenging. function designed simplify tasks, enabling seamless integration MGnify data HoloFood data retrieval database. need input returned data MGnifyR::getResult() HoloFoodR::getResult() functions.","code":""},{"path":"/reference/addMGnify.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Add results from MGnifyR to HoloFoodR results — addMGnify","text":"","code":"if (FALSE) { # \\dontrun{ # Get data from HoloFood database mae <- HoloFoodR::getResult(     salmon_sample_ids,     use.cache = TRUE )  # Get data from MGnify database mg <- MgnifyClient(     useCache = TRUE,     cacheDir = \".MGnifyR_cache\" ) tse <- MGnifyR::getResult(     mg,     accession = mgnify_analyses_ids,     get.func = FALSE )  # Add MGnify data to HoloFood data mae <- addMGnify(tse, mae) } # }"},{"path":"/reference/doQuery.html","id":null,"dir":"Reference","previous_headings":"","what":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"Search HoloFood database animals, genome catalogues, samples, viral catalogues","code":""},{"path":"/reference/doQuery.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"","code":"doQuery(type, flatten = TRUE, ...)"},{"path":"/reference/doQuery.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"type Character scalar specifying type data query. Must one following options: \"animals\", \"genome-catalogues\", \"samples\" \"viral-catalogues\". flatten Logical scalar specifying whether flatten resulting data.frame. means columns multiple values separated multiple columns. (Default: TRUE) ... optional arguments: max.hits NULL integer scalar specifying   maximum number results fetch. NULL, results fetched.   (Default: NULL) spread.sample.types Logical scalar specifying whether   create spread sample types column animals data. animals data,   sample types column might multiple values might hard   explore. argument specifies whether create presence/absence table   sample types. (Default: TRUE) use.cache Logical scalar specifying whether   use cache. (Default: FALSE) cache.dir Character scalar specifying cache   directory. (Default: tempdir()) clear.cache Logical scalar specifying whether   remove clear cache (Default: FALSE)","code":""},{"path":"/reference/doQuery.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"data.frame","code":""},{"path":"/reference/doQuery.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"doQuery flexible query function can utilized search available animals, genome catalogues, samples, viral catalogues. Search results can filtered; example, animals can filtered based available samples. See [Api browser](https://www.holofooddata.org/api/docs) information filters. can find help customizing queries [](https://emg-docs.readthedocs.io/en/latest/api.html#customising-queries).","code":""},{"path":"/reference/doQuery.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Search HoloFood database for animals, genome catalogues, samples, or viral catalogues — doQuery","text":"","code":"# Find animals results. The maximum amount of results is 100. Use filter # so that only chicken is searched. res <- doQuery(\"animals\", max.hits = 100, system = \"chicken\") head(res) #>        accession  system                                   canonical_url #> 1 SAMEA112904733 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904733 #> 2 SAMEA112904734 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904734 #> 3 SAMEA112904735 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904735 #> 4 SAMEA112904736 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904736 #> 5 SAMEA112904737 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904737 #> 6 SAMEA112904738 chicken https://www.ebi.ac.uk/biosamples/SAMEA112904738 #>   histological host_genomic inflammatory_markers metabolomic #> 1        FALSE         TRUE                FALSE       FALSE #> 2         TRUE         TRUE                 TRUE       FALSE #> 3         TRUE         TRUE                 TRUE       FALSE #> 4        FALSE         TRUE                FALSE       FALSE #> 5         TRUE         TRUE                 TRUE       FALSE #> 6        FALSE         TRUE                FALSE       FALSE #>   metabolomic_targeted metagenomic_assembly metatranscriptomic transcriptomic #> 1                FALSE                 TRUE              FALSE          FALSE #> 2                 TRUE                 TRUE               TRUE           TRUE #> 3                 TRUE                 TRUE              FALSE           TRUE #> 4                FALSE                 TRUE              FALSE          FALSE #> 5                 TRUE                 TRUE               TRUE           TRUE #> 6                 TRUE                 TRUE              FALSE          FALSE"},{"path":"/reference/getData.html","id":null,"dir":"Reference","previous_headings":"","what":"Get data from HoloFood database — getData","title":"Get data from HoloFood database — getData","text":"Get data HoloFood database","code":""},{"path":"/reference/getData.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get data from HoloFood database — getData","text":"","code":"getData(   type = NULL,   accession.type = NULL,   accession = NULL,   flatten = FALSE,   ... )"},{"path":"/reference/getData.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get data from HoloFood database — getData","text":"type NULL character scalar specifying type data query. Must one following options: \"analysis-summaries\", \"animals\", \"genome-catalogues\", \"samples\", \"sample_metadata_markers\" \"viral-catalogues\". genome viral catalogues fetched accession ID, type can also \"genomes\" \"fragments\". (Default: NULL) accession.type NULL character scalar specifying type accession IDs. Must one following options: \"animals\", \"genome-catalogues\", \"samples\" \"viral-catalogues\". (Default: NULL) accession NULL character vector specifying accession IDs type accession.type. (Default: NULL) flatten Logical scalar specifying whether flatten resulting data.frame. means columns multiple values separated multiple columns. (Default: FALSE) ... optional arguments: max.hits NULL integer scalar specifying   maximum number results fetch. NULL, results fetched.   (Default: NULL) use.cache Logical scalar specifying whether   use cache (Default: FALSE) cache.dir Character scalar specifying cache   directory. (Default: tempdir()) clear.cache Logical scalar specifying whether   remove clear cache (Default: FALSE)","code":""},{"path":"/reference/getData.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get data from HoloFood database — getData","text":"list data.frame","code":""},{"path":"/reference/getData.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get data from HoloFood database — getData","text":"getData, can fetch data database. Compared getResult, function flexible since can fetch kind data database. However, function returns data without wrangling list data.frame optimized format fetching data samples. Search results can filtered; example, animals can filtered based available samples. See [Api browser](https://www.holofooddata.org/api/docs) information filters. can find help customizing queries [](https://emg-docs.readthedocs.io/en/latest/api.html#customising-queries).","code":""},{"path":[]},{"path":"/reference/getData.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get data from HoloFood database — getData","text":"","code":"# Find genome catalogues catalogues <- getData(type = \"genome-catalogues\") head(catalogues) #>                 id                   title #> 1  salmon-gut-v2-0  HoloFood Salmon Gut v2 #> 2 chicken-gut-v2-0 HoloFood Chicken Gut v2 #>                                         biome related_mag_catalogue_id  system #> 1  root:Host-associated:Fish:Digestive system  non-model-fish-gut-v2-0  salmon #> 2 root:Host-associated:Birds:Digestive system       chicken-gut-v1-0-1 chicken #>   analysis_summaries #> 1       c(\"HoloF.... #> 2       c(\"HoloF....  # Find genomes based on certain genome catalogue iD res <- getData(     type = \"genomes\", accession.type = \"genome-catalogues\",     accession = catalogues[1, \"id\"], max.hits = 100) # See the data. head(res) #>       accession cluster_representative #> 1 MGYG000307500          MGYG000307500 #> 2 MGYG000307501          MGYG000307501 #> 3 MGYG000307502          MGYG000307501 #> 4 MGYG000307503          MGYG000299622 #> 5 MGYG000307504          MGYG000299579 #> 6 MGYG000307505          MGYG000299579 #>                                                                                                                      taxonomy #> 1                                                    Bacteria > Firmicutes_A > Clostridia > Oscillospirales > Ruminococcaceae #> 2 Bacteria > Proteobacteria > Gammaproteobacteria > Pseudomonadales > Pseudomonadaceae > Pseudomonas > Pseudomonas aeruginosa #> 3 Bacteria > Proteobacteria > Gammaproteobacteria > Pseudomonadales > Pseudomonadaceae > Pseudomonas > Pseudomonas aeruginosa #> 4          Bacteria > Proteobacteria > Gammaproteobacteria > Enterobacterales > Aeromonadaceae > Aeromonas > Aeromonas sobria #> 5                                                       Bacteria > Firmicutes > Bacilli > Mycoplasmatales > Mycoplasmoidaceae #> 6                                                       Bacteria > Firmicutes > Bacilli > Mycoplasmatales > Mycoplasmoidaceae #>                                                representative_url metadata1 #> 1 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000307500       196 #> 2 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000307501       197 #> 3 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000307501       198 #> 4 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000299622       199 #> 5 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000299579       200 #> 6 https://www.ebi.ac.uk/metagenomics/api/v1/genomes/MGYG000299579       201 #>   metadata.Genome_type metadata.Length metadata.N_contigs metadata.N50 #> 1                  MAG         1319783                 19       153761 #> 2                  MAG         6676386                136        83355 #> 3                  MAG         4766025               4741         1104 #> 4                  MAG         2381842                367         8519 #> 5                  MAG          681326                 38        37281 #> 6                  MAG          627706                 26        43119 #>   metadata.GC_content metadata.Completeness metadata.Contamination #> 1                27.3                 85.79                    0.0 #> 2               66.27                 98.37                  0.192 #> 3               64.82                 68.36                  2.053 #> 4               59.04                 58.95                   0.09 #> 5               25.14                 96.02                  0.384 #> 6               25.04                 95.25                  0.384 #>   metadata.rRNA_5S metadata.rRNA_16S metadata.rRNA_23S metadata.tRNAs #> 1              0.0               0.0               0.0             18 #> 2              0.0               0.0               0.0             17 #> 3            93.28              36.2             99.04             15 #> 4             91.6               0.0               0.0             14 #> 5              0.0             99.74             99.59             19 #> 6              0.0             99.74             99.59             19 #>   metadata.Genome_accession metadata.Sample_accession metadata.Study_accession #> 1               ERZ15182294            SAMEA112246717                ERP136460 #> 2               ERZ15182298            SAMEA112246686                ERP136460 #> 3               ERZ15182309            SAMEA112246683                ERP136460 #> 4               ERZ15182314            SAMEA112246713                ERP136460 #> 5               ERZ15233654            SAMEA112264417                ERP125469 #> 6               ERZ15233655            SAMEA112264472                ERP125469 #>   metadata.Country metadata.Continent #> 1           Norway             Europe #> 2           Norway             Europe #> 3           Norway             Europe #> 4           Norway             Europe #> 5           Norway             Europe #> 6           Norway             Europe #>                                                                                                                                       metadata.FTP_download #> 1 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0003075/MGYG000307500/genomes1/MGYG000307500.gff.gz #> 2 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0003075/MGYG000307501/genomes1/MGYG000307501.gff.gz #> 3 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0003075/MGYG000307501/genomes1/MGYG000307502.gff.gz #> 4 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0002996/MGYG000299622/genomes1/MGYG000307503.gff.gz #> 5 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0002995/MGYG000299579/genomes1/MGYG000307504.gff.gz #> 6 ftp://ftp.ebi.ac.uk/pub/databases/metagenomics/mgnify_genomes/non-model-fish-gut/v2.0/all_genomes/MGYG0002995/MGYG000299579/genomes1/MGYG000307505.gff.gz #>   metadata.Host_species annotations.cazy.GH annotations.cazy.PL #> 1           Salmo salar                   7                   0 #> 2           Salmo salar                  24                   0 #> 3           Salmo salar                  24                   0 #> 4           Salmo salar                  27                   0 #> 5           Salmo salar                   0                   0 #> 6           Salmo salar                   0                   0 #>   annotations.cazy.CE annotations.cazy.AA annotations.cazy.CB #> 1                   0                   0                   1 #> 2                   3                   1                   4 #> 3                   3                   1                   4 #> 4                   1                   1                   6 #> 5                   0                   0                   0 #> 6                   0                   0                   0 #>   annotations.cazy.GT annotations.cazy.CL #> 1                   7                   0 #> 2                  34                   0 #> 3                  34                   0 #> 4                  20                   0 #> 5                   2                   0 #> 6                   2                   0 # It includes for instance summary of the CAZy # (Carbohydrate-Active enZymes) annotations as a counts per category cazy <- res[ , grepl(\"annotations.cazy\", colnames(res)), drop = FALSE] head(cazy) #>   annotations.cazy.GH annotations.cazy.PL annotations.cazy.CE #> 1                   7                   0                   0 #> 2                  24                   0                   3 #> 3                  24                   0                   3 #> 4                  27                   0                   1 #> 5                   0                   0                   0 #> 6                   0                   0                   0 #>   annotations.cazy.AA annotations.cazy.CB annotations.cazy.GT #> 1                   0                   1                   7 #> 2                   1                   4                  34 #> 3                   1                   4                  34 #> 4                   1                   6                  20 #> 5                   0                   0                   2 #> 6                   0                   0                   2 #>   annotations.cazy.CL #> 1                   0 #> 2                   0 #> 3                   0 #> 4                   0 #> 5                   0 #> 6                   0 # Moreover, it includes a sample list. This sample list represents a  # collection of samples where the MAG was identified. Thr data has also the # completeness of MAG in a sample. head(res[ c(\"metadata.Sample_accession\", \"metadata.Completeness\")]) #>   metadata.Sample_accession metadata.Completeness #> 1            SAMEA112246717                 85.79 #> 2            SAMEA112246686                 98.37 #> 3            SAMEA112246683                 68.36 #> 4            SAMEA112246713                 58.95 #> 5            SAMEA112264417                 96.02 #> 6            SAMEA112264472                 95.25"},{"path":"/reference/getMetaboLights.html","id":null,"dir":"Reference","previous_headings":"","what":"Get metabolomic data from MetaboLights database — getMetaboLights","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"Get metabolomic data MetaboLights database","code":""},{"path":"/reference/getMetaboLights.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"","code":"getMetaboLights(study.id, ...)  getMetaboLightsFile(study.id, file, ...)"},{"path":"/reference/getMetaboLights.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"study.id character vector specifying study identifier data going fetched MetaboLights database. ... optional arguments: cache.dir Character scalar specifying directory   downloaded file stored. (Default: tempdir()) timeout Integer scalar specifying timeout   seconds loading file. (Default: 5*60) file character vector specifying files fetched.","code":""},{"path":"/reference/getMetaboLights.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"list","code":""},{"path":"/reference/getMetaboLights.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"HoloFood database primarily comprises targeted metabolomic data, omitting non-targeted metabolomic information. Nonetheless, features URLs linking studies within MetaboLights database. functionality enables users access non-targeted metabolomic data. getMetaboLights function returns structured list encompassing processed data data.frame format study metadata, assay metadata, assay. metadata includes file names spectra data. files can loaded getMetaboLightsFile. Alternatively, identified study files fetch, can refer [vignette](https://rformassspectrometry.github.io/MsIO/articles/MsIO.html#loading-data--metabolights) instructions loading data directly MsExperiment object, specifically designed metabolomics spectra data.","code":""},{"path":[]},{"path":"/reference/getMetaboLights.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get metabolomic data from MetaboLights database — getMetaboLights","text":"","code":"# This example is not run, because the server fails to respond sometimes. if( FALSE ){     res <- getMetaboLights(\"MTBLS4381\")     file_paths <- getMetaLightsFile(         study.id = \"MTBLS4381\",         file = res[[\"assay_meta\"]][[\"Raw Spectral Data File\"]]         ) }"},{"path":"/reference/getResult.html","id":null,"dir":"Reference","previous_headings":"","what":"Get data on samples from HoloFood database — getResult","title":"Get data on samples from HoloFood database — getResult","text":"Get data samples HoloFood database","code":""},{"path":"/reference/getResult.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Get data on samples from HoloFood database — getResult","text":"","code":"getResult(accession, ...)"},{"path":"/reference/getResult.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Get data on samples from HoloFood database — getResult","text":"accession Character vector specifying accession IDs type samples. ... optional arguments: use.cache Logical scalar specifying whether   use cache. Note get.metabolomic = TRUE specified,   file MetaboLights stored local system location   specified cache.dir despite value use.cache.   (Default: FALSE) cache.dir Character scalar specifying cache   directory. (Default: tempdir()) clear.cache Logical scalar specifying whether   use.cache (Default: FALSE) assay.type Character scalar specifying name   assay resulting TreeSummarizedExperiment object.   (Default: \"counts\") get.metabolomic Logical scalar specifying whether   retrieve processed metabolomic data MetaboLights database.   retrieving spectra data, refer   getMetaboLights documentation.   (Default: FALSE)","code":""},{"path":"/reference/getResult.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Get data on samples from HoloFood database — getResult","text":"MultiAssayExperiment","code":""},{"path":"/reference/getResult.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Get data on samples from HoloFood database — getResult","text":"getResult, can fetch data samples HoloFood database. Compared getData, function convenient fetching samples data converts data MultiAssayExperiment different omics stored TreeSummarizedExperiment objects optimized downstream analytics. Columns returned MultiAssayExperiment individual animals. columns linked individual samples stored TreeSummarizedExperiment objects. HoloFood database lacks non-targeted metabolomic data can fetched MetaboLights resource. Certain datasets include processed features. datasets can retrieved function getResult integrates metabolomic data datasets HoloFood. Furthermore, HoloFoodR database include metagenomic assembly data, users can access data MGnify database. MGnifyR package provides convenient interface accessing database. employing MGnifyR::getResult(), users can obtain data formatted MultiAssayExperiment object, containing multiple TreeSummarizedExperiment objects. Consequently, data HoloFood MGnify databases inherently compatible subsequent downstream analysis.","code":""},{"path":[]},{"path":"/reference/getResult.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Get data on samples from HoloFood database — getResult","text":"","code":"# Find samples on certain animal samples <- doQuery(\"samples\", animal_accession = \"SAMEA112904746\")  # Get the data mae <- getResult(samples[[\"accession\"]]) #> Warning: Data for the following samples cannot be found. The sample types are metagenomic_assembly, host_genomic, transcriptomic and metatranscriptomic. (Note that metagenomic assemblies can be found from the MGnify database. See MGnifyR package.): #> 'SAMEA10130039', 'SAMEA13389496', 'SAMEA13389497', 'SAMEA13901618', 'SAMEA13901619', 'SAMEA13929785', 'SAMEA7571815' mae #> A MultiAssayExperiment object of 8 listed #>  experiments with user-defined names and respective classes. #>  Containing an ExperimentList class object of length 8: #>  [1] BIOGENIC AMINES: TreeSummarizedExperiment with 7 rows and 2 columns #>  [2] FATTY ACIDS: TreeSummarizedExperiment with 19 rows and 2 columns #>  [3] HISTOLOGY: TreeSummarizedExperiment with 20 rows and 2 columns #>  [4] INFLAMMATORY MARKERS: TreeSummarizedExperiment with 14 rows and 2 columns #>  [5] metagenomic_assembly: TreeSummarizedExperiment with 0 rows and 2 columns #>  [6] host_genomic: TreeSummarizedExperiment with 0 rows and 2 columns #>  [7] transcriptomic: TreeSummarizedExperiment with 0 rows and 2 columns #>  [8] metatranscriptomic: TreeSummarizedExperiment with 0 rows and 1 columns #> Functionality: #>  experiments() - obtain the ExperimentList instance #>  colData() - the primary/phenotype DataFrame #>  sampleMap() - the sample coordination DataFrame #>  `$`, `[`, `[[` - extract colData columns, subset, or experiment #>  *Format() - convert into a long or wide DataFrame #>  assays() - convert ExperimentList to a SimpleList of matrices #>  exportClass() - save data to flat files"}]