Added colour option to cmpb pipeline #8

draeger-lab · May 24, 2024 · 5465e0e · 5465e0e
1 parent 1f4e3af
commit 5465e0e
Show file tree

Hide file tree

Showing 3 changed files with 21 additions and 68 deletions.
diff --git a/dev/CB_test_modules.ipynb b/dev/CB_test_modules.ipynb
@@ -2769,7 +2769,7 @@
     {
      "data": {
       "text/plain": [
-       "<bound method SBase.getNotesString of <Model Kp_std \"Genome-scale metabolic model of K. pneumoniae HS11286\">>"
+       "'test'"
       ]
      },
      "execution_count": 2,
@@ -2778,63 +2778,10 @@
     }
    ],
    "source": [
-    "from refinegems.utility.io import load_model\n",
+    "import cobra\n",
     "\n",
-    "p = '/Users/brune/Documents/11_Test_Data/test_SPECIMEN/thesis/Kp_std/02_generate_draft_model/Kp_std_draft.xml'\n",
-    "model = load_model(p,'libsbml')\n",
-    "\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "model.getModelHistory()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "'<notes>\\n  <html xmlns=\"http://www.w3.org/1999/xhtml\">\\n    <p>Description: This model was built with CarveMe version 1.5.1</p>\\n  </html>\\n</notes>'"
-      ]
-     },
-     "execution_count": 24,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "x = model.getNotesString()\n",
-    "x\n",
-    "# 'CarveMe' in x"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'Description': 'This model was built with CarveMe version 1.5.1'}"
-      ]
-     },
-     "execution_count": 13,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "mod2 = load_model(p, 'cobra')\n",
-    "mod2.notes"
+    "m = cobra.Model('test')\n",
+    "m.id"
    ]
   },
   {

diff --git a/src/specimen/cmpb/workflow.py b/src/specimen/cmpb/workflow.py
@@ -40,9 +40,6 @@
 
     # ....................................................
     # @TODO / @IDEAS
-    # global options
-    # run memote after every step
-    # calculate model stats after each step
     # use temp folder or report all model/in-between steps
     # what to write in the log file
     # ....................................................
@@ -51,9 +48,6 @@
 #   in the run function: current_model means the cobrapy model, 
 #   while current_libmodel means the libsbml model
 
-# @TODO / @IDEAS Add option to have specific colour list per model for plots
-# @TODO Maybe get models at first and then add model IDs to every save filename?
-
 def run(configpath:str):
 
     def between_growth_test(model, cfg, step):
@@ -66,7 +60,8 @@ def between_growth_test(model, cfg, step):
             growth_report = growth.growth_analysis(model, cfg['input']['mediapath'], 
                                                 namespace=cfg['general']['namespace'], 
                                                 retrieve='report')
-            growth_report.save(Path(cfg['general']["dir"], 'cmpb_out', 'misc', 'growth', step)) 
+            growth_report.save(Path(cfg['general']["dir"], 'cmpb_out', 'misc', 'growth', step),
+                               color_palette=config['general']['colours']) 
         else:
             mes = f'No growth/biomass function detected, growth simulation for step {step} will be skipped.'
             warnings.warn(mes)
@@ -80,7 +75,8 @@ def between_analysis(model, cfg, step):
                     verbose=False)
         if cfg['general']['stats_always_on']:
             report = ModelInfoReport(model)
-            report.save(Path(cfg['general']["dir"],'cmpb_out', 'misc', 'stats',f'{step}_.html')) 
+            report.save(Path(cfg['general']["dir"],'cmpb_out', 'misc', 'stats',f'{step}_.html'),
+                        color_palette=config['general']['colours']) 
 
 
     # setup phase
@@ -374,12 +370,14 @@ def between_analysis(model, cfg, step):
     # stats
     # -----
     stats_report = ModelInfoReport(current_model)
-    stats_report.save(Path(dir,'cmpb_out', 'misc','stats')) 
+    stats_report.save(Path(dir,'cmpb_out', 'misc','stats'),
+                      color_palette=config['general']['colours']) 
 
     # kegg pathway
     # ------------
     pathway_report = kegg_pathway_analysis(current_model)
-    pathway_report.save(Path(dir,'cmpb_out','misc','kegg_pathway')) 
+    pathway_report.save(Path(dir,'cmpb_out','misc','kegg_pathway'), 
+                        colors=config['general']['colours']) 
 
     # sbo term
     # --------
@@ -398,11 +396,16 @@ def between_analysis(model, cfg, step):
     # ------------
     media_list = growth.read_media_config(config['input']['mediapath'])
     auxo_report = growth.test_auxotrophies(current_model, media_list[0], media_list[1], config['general']['namespace'])
-    auxo_report.save(Path(dir,'cmpb_out','misc','auxotrophies')) 
+    auxo_report.save(Path(dir,'cmpb_out','misc','auxotrophies'),
+                     color_palette=config['general']['colours']) 
 
 
 ####### IDEAS below ####
 
+# @TODO / @IDEAS Add option to have specific colour list per model for plots
+    # for the comparison when runnin this on multiple models
+# @TODO Maybe get models at first and then add model IDs to every save filename?
+
 # run for multiple models
 def wrapper():
     pass
diff --git a/src/specimen/data/config/cmpb_config.yaml b/src/specimen/data/config/cmpb_config.yaml
@@ -22,11 +22,14 @@ input:
 # ---------------
 general:
   dir: './'                  # Path/Name of a directory to save output to
+  colours: 'YlGn'            # set the colour scheme for the plots
+                             # should be a valid matplotlib continuous color palette
   namespace: BiGG            # Namespace to use for the model
                              # Possible identifiers, currently: BiGG
   save_all_models: True      # save all models (models for each step)
   memote_always_on: False    # run memote after every step
   stats_always_on: False     # calculate the model statistics after every step
+  # below are options used by multiple steps
   refseq_gff: USER           # Path to RefSeq GFF file: Required for gap analysis with 'KEGG'. 
                              # Can be optionally provided for cm-polish.
   kegg_organism_id: USER     # KEGG ID of the organism: Required for gap analysis with KEGG.