missing features bugfix, deeplc logging

compomics · Aug 21, 2023 · 5599e44 · 5599e44
1 parent dfa0c93
commit 5599e44
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 7 deletions.
diff --git a/ms2rescore/core.py b/ms2rescore/core.py
@@ -105,19 +105,19 @@ def rescore(configuration: Dict) -> None:
         (set(psm.rescoring_features.keys()) == all_feature_names) for psm in psm_list
     ]
 
-    psm_list = psm_list[psms_with_features]
-
     if psms_with_features.count(False) > 0:
+        removed_psms = psm_list[[not psm for psm in psms_with_features]]
         missing_features = {
             feature_name
-            for present_features in psm_list[[not presence for presence in psms_with_features]]
-            for feature_name in (all_feature_names - present_features)
+            for psm in removed_psms
+            for feature_name in all_feature_names - set(psm.rescoring_features.keys())
         }
         logger.warning(
-            f"Removed {psms_with_features.count(False)} PSMs that were missing one or more"
-            f"rescoring feature, {missing_features}."
+            f"Removed {psms_with_features.count(False)} PSMs that were missing one or more "
+            f"rescoring feature(s), {missing_features}."
         )
 
+    psm_list = psm_list[psms_with_features]
     # Write feature names to file
     _write_feature_names(feature_names, output_file_root)
 

diff --git a/ms2rescore/feature_generators/deeplc.py b/ms2rescore/feature_generators/deeplc.py
@@ -111,12 +111,16 @@ def add_features(self, psm_list: PSMList) -> None:
         )
 
         # Run MS²PIP for each spectrum file
+        total_runs = len(psm_dict.keys())
+        current_run = 1
         for runs in psm_dict.values():
             # Reset DeepLC predictor for each collection of runs
             self.deeplc_predictor = None
             self.selected_model = None
             for run, psms in runs.items():
-                logger.info(f"Running DeepLC for PSMs from run `{run}`...")
+                logger.info(
+                    f"Running DeepLC for PSMs from run ({current_run}/{total_runs}): `{run}`..."
+                )
                 # Prepare PSM file
                 with contextlib.redirect_stdout(
                     open(os.devnull, "w")
@@ -182,6 +186,7 @@ def add_features(self, psm_list: PSMList) -> None:
                                 "predicted_retention_time_best": predictions[i],
                                 "rt_diff_best": rt_diffs_run[i],
                             }
+            current_run += 1
 
         for psm in psm_list:
             psm["rescoring_features"].update(