Use list of gene symbols and set nan values to None as before

ci/tests/test_transcriptformer/test_transcriptformer_model.py

@@ -22,11 +22,11 @@ def test_process_data__correct_ensembl_ids(self):
         dataset = self.transcriptformer.process_data([self.data])
         assert len(dataset) == 1
         assert all(
-            dataset.files_list[0].var["ensembl_id"].astype(str).values
+            dataset.files_list[0].var["ensembl_id"].values
             == [
                 "ENSG00000121410",
                 "ENSG00000036549",
-                "nan",
+                None,
                 "ENSG00000074755",
             ]
         )

helical/utils/mapping.py

@@ -105,7 +105,7 @@ def convert_list_ensembl_ids_to_gene_symbols(
     """
     df = _get_ensembl_mart_df(species=species)
     mapping = df.drop_duplicates(subset="ensembl_id").set_index("ensembl_id")["gene_name"]
-    return list(pd.Series(ensembl_ids, dtype="object").map(mapping))
+    return list(pd.Series(list(ensembl_ids), dtype="object").map(mapping).where(pd.notna, None))
 
 
 def convert_list_gene_symbols_to_ensembl_ids(
@@ -128,4 +128,4 @@ def convert_list_gene_symbols_to_ensembl_ids(
     """
     df = _get_ensembl_mart_df(species=species)
     mapping = df.drop_duplicates(subset="gene_name").set_index("gene_name")["ensembl_id"]
-    return list(pd.Series(gene_symbols, dtype="object").map(mapping))
+    return list(pd.Series(list(gene_symbols), dtype="object").map(mapping).where(pd.notna, None))

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "helical"
-version = "1.4.8"
+version = "1.4.9"
 authors = [
   { name="Helical Team", email="support@helical-ai.com" },
 ]