Merge pull request #556 from pavanvidem/scrna-scanpy

Add a new scRNA workflow for standard analysis using Scanpy

workflows/scRNAseq/scanpy-clustering/.dockstore.yml

@@ -0,0 +1,17 @@
+version: 1.2
+workflows:
+- name: main
+  subclass: Galaxy
+  publish: true
+  primaryDescriptorPath: /Preprocessing-and-Clustering-of-single-cell-RNA-seq-data-with-Scanpy.ga
+  testParameterFiles:
+  - /Preprocessing-and-Clustering-of-single-cell-RNA-seq-data-with-Scanpy-tests.yml
+  authors:
+  - name: Pavankumar Videm
+    orcid: 0000-0002-5192-126X
+  - name: Hans-Rudolf Hotz
+    orcid: 0000-0002-2799-424X
+  - name: Mehmet Tekman
+    orcid: 0000-0002-4181-2676
+  - name: "Bérénice Batut"
+    orcid: 0000-0001-9852-1987

workflows/scRNAseq/scanpy-clustering/CHANGELOG.md

@@ -0,0 +1,3 @@
+## [0.1] 2024-10-09
+
+First release.

workflows/scRNAseq/scanpy-clustering/Preprocessing-and-Clustering-of-single-cell-RNA-seq-data-with-Scanpy-tests.yml

@@ -0,0 +1,205 @@
+- doc: Test outline for Preprocessing-and-Clustering-of-single-cell-RNA-seq-data-with-Scanpy
+  job:
+    Barcodes:
+      class: File
+      location: https://zenodo.org/record/3581213/files/barcodes.tsv
+      filetype: txt
+    Genes:
+      class: File
+      location: https://zenodo.org/record/3581213/files/genes.tsv
+      filetype: tabular
+    Matrix:
+      class: File
+      location: https://zenodo.org/record/3581213/files/matrix.mtx
+      filetype: mtx
+    Minimum number of cells expressed: 3
+    Input is from Cell Ranger v2 or earlier versions: true
+    Mitochondrial genes start with pattern: 'MT-'
+    Minimum number of genes expressed: 200
+    Maximum number of genes expressed: 2500
+    Number of neighbours for computing neighborhood graph: 10
+    Number of PCs to use for computing neighborhood graph: 10
+    Louvain resolution: 0.45
+    Manually annotate celltypes?: true
+    Annotate louvain clusters with these cell types: 'CD4+ T, CD14+, B, CD8+ T, FCGR3A+,
+      NK, Dendritic, Megakaryocytes'
+  outputs:
+    Anndata with Celltype Annotation:
+      asserts:
+        has_h5_keys:
+            keys: "obs/louvain"
+            keys: "var/highly_variable"
+            keys: "uns/rank_genes_groups"
+    UMAP of louvain and top ranked genes:
+      has_size:
+        size: 696399
+        delta: 60000
+      has_image_width:
+        width: 1146
+        delta: 60
+      has_image_height:
+        height: 2099
+        delta: 105
+    Dotplot of top genes on clusters:
+      has_size:
+        size: 39916
+        delta: 3900
+      has_image_width:
+        width: 535
+        delta: 27
+      has_image_height:
+        height: 379
+        delta: 19
+    Initial Anndata General Info:
+      asserts:
+        has_text:
+            text: "AnnData object with n_obs × n_vars = 2700 × 32738"
+    Scatter plot of plot n_counts vs n_genes:
+      has_size:
+        size: 27381
+        delta: 2700
+      has_image_width:
+        width: 630
+        delta: 40
+      has_image_height:
+        height: 413
+        delta: 25
+    Plot highly variable:
+      has_size:
+        size: 67135
+        delta: 6500
+      has_image_width:
+        width: 1061
+        delta: 50
+      has_image_height:
+        height: 432
+        delta: 27
+    Plot PCA over view with genes:
+      has_size:
+        size: 61974
+        delta: 600
+      has_image_width:
+        width: 1158
+        delta: 60
+      has_image_height:
+        height: 409
+        delta: 25
+    UMAP of louvain:
+      has_size:
+        size: 68416
+        delta: 6000
+      has_image_width:
+        width: 601
+        delta: 30
+      has_image_height:
+        height: 429
+        delta: 25
+    Ranked genes with Wilcoxon test:
+      asserts:
+        has_line:
+            line: "LDHB	LYZ	CD74	CCL5	LST1	NKG7	HLA-DPA1	PF4"
+    Plot Rank gene groups:
+      has_size:
+        size: 164953
+        delta: 16000
+      has_image_width:
+        width: 1559
+        delta: 75
+      has_image_height:
+        height: 1192
+        delta: 50
+    UMAP with annotated cell types:
+      has_size:
+        size: 70265
+        delta: 7000
+      has_image_width:
+        width: 535
+        delta: 27
+      has_image_height:
+        height: 429
+        delta: 25
+    Stacked violin of top genes on clusters:
+      has_size:
+        size: 22147
+        delta: 2200
+      has_image_width:
+        width: 299
+        delta: 15
+      has_image_height:
+        height: 258
+        delta: 15
+    Heatmap of top 20 highly ranked genes:
+      has_size:
+        size: 162549
+        delta: 16000
+      has_image_width:
+        width: 788
+        delta: 40
+      has_image_height:
+        height: 505
+        delta: 25
+    Scatter plot of n_genes vs pct_mito:
+      has_size:
+        size: 33175
+      has_image_width:
+        width: 617
+        delta: 30
+      has_image_height:
+        height: 413
+        delta: 20
+    Violin plot n_genes, n_counts, pct_mito:
+      has_size:
+        size: 71188
+      has_image_width:
+        width: 1508
+        delta: 75
+      has_image_height:
+        height: 490
+        delta: 25
+    Anndata with raw attribute:
+      asserts:
+        has_h5_keys:
+            keys: "obs/log1p_total_counts"
+            keys: "obs/total_counts_mito"
+            keys: "var/mito"
+            keys: "var/norm"
+            keys: "uns/log1p"
+    Plot genes according to contributions to PCs:
+      has_size:
+        size: 71020
+        delta: 7000
+      has_image_width:
+        width: 1601
+        delta: 80
+      has_image_height:
+        height: 430
+        delta: 27
+    Elbow plot of PCs and variance:
+      has_size:
+        size: 20174
+        delta: 2000
+      has_image_width:
+        width: 569
+        delta: 25
+      has_image_height:
+        height: 453
+        delta: 25
+    General information about the final Anndata object:
+      path: test-data/General information about the final Anndata object.txt
+
+    Number of cells per cluster:
+      asserts:
+        has_line:
+            line: "0	1162"
+            line: "3	311"
+            line: "6	34"
+    Violin plot of top genes on clusters:
+      has_size:
+        size: 261433
+        delta: 26000
+      has_image_width:
+        width: 5532
+        delta: 280
+      has_image_height:
+        height: 418
+        delta: 20