Fix to possible_replacement so the list field is standardized after i…

…t's actually a list of strings
Sage-Bionetworks · Nov 22, 2024 · 3a60655 · 3a60655
1 parent ab1bb82
commit 3a60655
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/data_analysis/agora/notebooks/preprocessing/AG-896_Preprocess_Gene_Annotations.ipynb b/data_analysis/agora/notebooks/preprocessing/AG-896_Preprocess_Gene_Annotations.ipynb
@@ -580,11 +580,11 @@
    "source": [
     "gene_table_merged[\"possible_replacement\"] = gene_table_merged[\n",
     "    \"possible_replacement\"\n",
-    "].apply(preprocessing_utils.standardize_list_item)\n",
+    "].apply(lambda pr: pr if pr is np.NaN or len(pr) == 0 else [x[\"stable_id\"] for x in pr])\n",
     "\n",
     "gene_table_merged[\"possible_replacement\"] = gene_table_merged[\n",
     "    \"possible_replacement\"\n",
-    "].apply(lambda pr: pr if len(pr) == 0 else [x[\"stable_id\"] for x in pr])\n",
+    "].apply(preprocessing_utils.standardize_list_item)\n",
     "\n",
     "gene_table_merged = gene_table_merged[\n",
     "    [\n",