fix: get_features_list_from_metadata returned cols that were not feat…

…ures
opentargets · Feb 13, 2025 · 58f35d9 · 58f35d9
1 parent c635e18
commit 58f35d9
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 2 deletions.
diff --git a/src/gentropy/dataset/l2g_prediction.py b/src/gentropy/dataset/l2g_prediction.py
@@ -96,7 +96,6 @@ def from_credible_set(
                     .join(feature_matrix._df, "studyLocusId")
                     .filter(f.col("isProteinCoding") == 1)
                 ),
-                features_list=l2g_model.features_list,
             )
             .fill_na()
             .select_features(l2g_model.features_list)

diff --git a/src/gentropy/method/l2g/model.py b/src/gentropy/method/l2g/model.py
@@ -136,7 +136,13 @@ def get_features_list_from_metadata() -> list[str]:
             return [
                 column
                 for column in model_config["sklearn"]["columns"]
-                if column != "studyLocusId"
+                if column
+                not in [
+                    "studyLocusId",
+                    "geneId",
+                    "traitFromSourceMappedId",
+                    "goldStandardSet",
+                ]
             ]
 
         local_path = model_id