update analysis to be compatible with > 3.9

README.md

@@ -15,7 +15,7 @@ Topyfic's full documentation can be found at [here](https://mortazavilab.github.
 
 ## Installation
 
-To install Topyfic, python version 3.8 or greater is required.
+To install Topyfic, python version 3.9 or greater is required.
 
 **IMPORTANT**: If you upgrade Topyfic from any version below 0.4.5 you need to run this
 

Topyfic/analysis.py

@@ -40,10 +40,10 @@ def __init__(self,
         if colors_topics is None:
             colors = sns.color_palette("turbo", self.top_model.N).as_hex()
 
-            def myfunction():
-                return 0.1
+            #def myfunction():
+            #    return 0.1
 
-            random.shuffle(colors, myfunction)
+            random.shuffle(colors)
             index = list(self.top_model.topics.keys())
             self.colors_topics = pd.DataFrame({'colors': colors}, index=index)
         else:

workflow/snakemake/workflow/Snakefile

@@ -23,7 +23,8 @@ include: "rules/topModel.smk"
 #### check if merging is necessary ###
 merging = []
 if 'merge' in config:
-    merging = f"{config['workdir']}/topModel_{'_'.join(config['names'])}.p"
+    merging = [f"{config['workdir']}/topModel_{'_'.join(config['names'])}.p",
+               f"{config['workdir']}/analysis_{'_'.join(config['names'])}.p"]
 
 
 ##### target rules #####
@@ -39,4 +40,7 @@ rule all:
         expand(f"{config['workdir']}/{{name}}/{{n_topic}}/topmodel/topModel_{{name}}_{{n_topic}}.p",
             name=config['names'],
             n_topic=config["n_topics"]),
+        expand(f"{config['workdir']}/{{name}}/{{n_topic}}/topmodel/analysis_{{name}}_{{n_topic}}.p",
+            name=config['names'],
+            n_topic=config["n_topics"]),
         merging,

workflow/snakemake/workflow/rules/topModel.smk

@@ -11,7 +11,7 @@ rule run_top_model:
             name=config["names"],
             n_topic=config["n_topics"])
     output:
-        f"{config['workdir']}/{{name}}/{{n_topic}}/topmodel/topModel_{{name}}_{{n_topic}}.p",
+        f"{config['workdir']}/{{name}}/{{n_topic}}/topmodel/topModel_{{name}}_{{n_topic}}.p"
     params:
         name=lambda wildcards: wildcards.name,
         n_topic=lambda wildcards: wildcards.n_topic,
@@ -32,6 +32,24 @@ rule run_top_model:
                 config['top_model']['min_cell_participation']),
             topmodel_output=f"{config['workdir']}/{params.name}/{params.n_topic}/topmodel/")
 
+# Rule to run analysis for one input adata
+rule run_analysis:
+    input:
+        expand(f"{config['workdir']}/{{name}}/{{n_topic}}/topmodel/topModel_{{name}}_{{n_topic}}.p",
+            name=config["names"],
+            n_topic=config["n_topics"])
+    output:
+        f"{config['workdir']}/{{name}}/{{n_topic}}/topmodel/analysis_{{name}}_{{n_topic}}.p"
+    params:
+        name=lambda wildcards: wildcards.name,
+        n_topic=lambda wildcards: wildcards.n_topic,
+    run:
+        top_model = Topyfic.read_topModel(f"{config['workdir']}/{params.name}/{params.n_topic}/topmodel/topModel_{params.name}_{params.n_topic}.p")
+
+        make_topmodel.make_analysis(adata_paths=config['count_adata'][params.name],
+            top_model=top_model,
+            topmodel_output=f"{config['workdir']}/{params.name}/{params.n_topic}/topmodel/")
+
 # Rule to run topModel for merging multiple input adatas
 if 'merge' in config:
     rule run_multiple_top_model:
@@ -75,3 +93,9 @@ if 'merge' in config:
                 min_cell_participation=None if config['top_model']['min_cell_participation'] == "None" else float(
                     config['top_model']['min_cell_participation']),
                 topmodel_output=f"{config['workdir']}")
+
+            top_model = Topyfic.read_topModel(f"{config['workdir']}/topModel_{'_'.join(config['names'])}.p")
+
+            make_topmodel.make_analysis(adata_paths=adata_paths,
+                top_model=top_model,
+                topmodel_output=f"{config['workdir']}")

workflow/snakemake/workflow/scripts/make_topmodel.py

@@ -30,13 +30,13 @@ def find_best_n_topic(n_topics, names, topmodel_outputs):
 
 def make_top_model(trains, adata_paths, n_top_genes=50, resolution=1, max_iter_harmony=10, min_cell_participation=None,
                    topmodel_output=None):
-
     # Check if the file exists
     if not os.path.exists(topmodel_output):
         os.makedirs(topmodel_output, mode=0o777)
 
+    sc.settings.figdir = f"{topmodel_output}/figures/"
+
     adata = None
-    print(adata_paths)
     if isinstance(adata_paths, str):
         adata = sc.read_h5ad(adata_paths)
     else:
@@ -56,10 +56,28 @@ def make_top_model(trains, adata_paths, n_top_genes=50, resolution=1, max_iter_h
 
     print(top_model.name, topmodel_output)
     top_model.save_topModel(save_path=topmodel_output)
+    print(f"{topmodel_output}topModel_{top_model.name}.p")
 
     adata_topmodel.write_h5ad(f"{topmodel_output}/topic_weight_umap.h5ad")
     clustering.to_csv(f"{topmodel_output}/topic_cluster_mapping.csv")
 
+
+def make_analysis(adata_paths, top_model, topmodel_output=None):
+    # Check if the file exists
+    if not os.path.exists(topmodel_output):
+        os.makedirs(topmodel_output, mode=0o777)
+
+    adata = None
+    if isinstance(adata_paths, str):
+        adata = sc.read_h5ad(adata_paths)
+    else:
+        for adata_path in adata_paths:
+            tmp = sc.read_h5ad(adata_path)
+            if adata is None:
+                adata = sc.read_h5ad(adata_path)
+            else:
+                adata = adata.concatenate(tmp)
+
     analysis_top_model = Topyfic.Analysis(Top_model=top_model)
     analysis_top_model.calculate_cell_participation(data=adata)
     analysis_top_model.save_analysis(save_path=topmodel_output)