fix bug in delimiter_in_gene_name and add test

Author: jlchang

ingest/de.py

@@ -224,8 +224,9 @@ def execute_de(self):
     @staticmethod
     def get_genes(genes_path):
         """ Genes file can have one or two columns of gene information
-            If two columns present, check if there are duplicates in 2nd col
-            If no duplicates, use as var_names, else use join columns
+            Preferentially use gene names from second column.
+            If duplicate gene names, check that 1st plus 2nd column provides uniqueness
+            If unique when joined, join columns with pipe (|) for use as DE input
         """
         genes_object = IngestFiles(genes_path, None)
         local_genes_path = genes_object.resolve_path(genes_path)[1]
@@ -302,7 +303,7 @@ def remove_single_sample_data(adata, annotation):
     def delimiter_in_gene_name(rank):
         """ Check if pipe delimiter occurs in "names" column
         """
-        return rank['names'].str.contains('|').any
+        return rank['names'].str.contains('|', regex=False).any()
 
     @staticmethod
     def extract_gene_id_for_out_file(rank):

tests/test_de.py

@@ -110,6 +110,24 @@ def test_detect_duplicate_gene_names(self):
         dup_genes = DifferentialExpression.get_genes(dup_genes_path)
         self.assertIn("|", dup_genes[0], f"no delimiter expected in {dup_genes[0]}")
 
+    def test_delimiter_in_gene_name(self):
+        delimited_data = {"names": ["Tns1", "Gfra1"], "scores": ["10.5", "10.34"]}
+        delimited_df = pd.DataFrame(delimited_data)
+        self.assertFalse(
+            DifferentialExpression.delimiter_in_gene_name(delimited_df),
+            "no pipe delimiter should be detected in the input",
+        )
+
+        undelimited_data = {
+            "names": ["ENSMUST00000027035|Sox17", "ENSMUST00000195555|Sox17"],
+            "scores": ["41.459137", "-5.058518"],
+        }
+        undelimited_df = pd.DataFrame(undelimited_data)
+        self.assertTrue(
+            DifferentialExpression.delimiter_in_gene_name(undelimited_df),
+            "expected pipe delimiter undetected",
+        )
+
     def test_de_remove_single_sample(self):
         """ Test single sample removal
         """