Merge branch 'gtf_fix'

colomemaria · Dec 12, 2023 · 0e845ef · 0e845ef
2 parents 070a946 + a2daf62
commit 0e845ef
Show file tree

Hide file tree

Showing 3 changed files with 19 additions and 9 deletions.
diff --git a/episcanpy/count_matrix/_bld_atac_mtx.py b/episcanpy/count_matrix/_bld_atac_mtx.py
@@ -394,8 +394,16 @@ def gene_activity_mtx(fragments_file,
         features = features[features.source == source]
 
     features["gene_id"] = [attr.replace("gene_id", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene_id")]
-    features["gene_name"] = [attr.replace("gene_name", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene_name")]
-    features["gene_type"] = [attr.replace("gene_type", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene_type")]
+
+    tmp = [attr.replace("gene_name", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene_name ")]
+    if not tmp:
+        tmp = [attr.replace("gene", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene ")]
+    features["gene_name"] = tmp
+
+    tmp = [attr.replace("gene_type", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene_type ")]
+    if not tmp:
+        tmp = [attr.replace("gene_biotype", "").strip().strip("\"") for feature_attr in features.attribute for attr in feature_attr.split(";") if attr.strip().startswith("gene_biotype ")]
+    features["gene_type"] = tmp
 
     if gene_type:
         features = features[[feature in gene_type for feature in features.gene_type]]

diff --git a/episcanpy/preprocessing/_quality_control.py b/episcanpy/preprocessing/_quality_control.py
@@ -851,24 +851,24 @@ def set_filter(adata,
         df = adata.obs if in_obs else adata.var
 
         if "passes_filter" in df:
-            if min_threshold and max_threshold:
+            if min_threshold is not None and max_threshold is not None:
                 tmp = np.logical_and(df[key] >= min_threshold, df[key] <= max_threshold)
                 df["passes_filter"] = [False if not passed else val for passed, val in zip(tmp, df["passes_filter"])]
-            elif min_threshold:
+            elif min_threshold is not None:
                 tmp = df[key] >= min_threshold
                 df["passes_filter"] = [False if not passed else val for passed, val in zip(tmp, df["passes_filter"])]
-            elif max_threshold:
+            elif max_threshold is not None:
                 tmp = df[key] <= max_threshold
                 df["passes_filter"] = [False if not passed else val for passed, val in zip(tmp, df["passes_filter"])]
 
         else:
-            if min_threshold and max_threshold:
+            if min_threshold is not None and max_threshold is not None:
                 tmp = np.logical_and(df[key] >= min_threshold, df[key] <= max_threshold)
                 df["passes_filter"] = tmp
-            elif min_threshold:
+            elif min_threshold is not None:
                 tmp = df[key] >= min_threshold
                 df["passes_filter"] = tmp
-            elif max_threshold:
+            elif max_threshold is not None:
                 tmp = df[key] <= max_threshold
                 df["passes_filter"] = tmp
 

diff --git a/episcanpy/preprocessing/_tss_enrichment.py b/episcanpy/preprocessing/_tss_enrichment.py
@@ -47,6 +47,7 @@ def get_tss(gtf,
 def tss_enrichment(adata,
                    fragments,
                    gtf,
+                   source="HAVANA",
                    n=5000,
                    score="avg_score_of_center_region",
                    distance_to_tss=1000,
@@ -59,6 +60,7 @@ def tss_enrichment(adata,
         adata: AnnData
         fragments: path to fragments file
         gtf: path to GTF file
+        source: source of the feature
         n: number of TSS to use for calculation
         score: value that is used as TSS enrichment score for individual observations
         distance_to_tss: distance to TSS
@@ -69,7 +71,7 @@ def tss_enrichment(adata,
         None
     """
 
-    features = get_tss(gtf, source="HAVANA", feature="gene", protein_coding_only=True)
+    features = get_tss(gtf, source=source, feature="gene", protein_coding_only=True)
 
     features["start"] = features.tss_pos - distance_to_tss
     features["stop"] = features.tss_pos + distance_to_tss