Merge pull request #138 from compomics/ms2pip-fgen-improvements

Various minor improvements

ms2rescore/core.py

@@ -157,7 +157,7 @@ def rescore(configuration: Dict, psm_list: Optional[PSMList] = None) -> None:
                 output_file_root, psm_list=psm_list, feature_names=feature_names, use_txt_log=True
             )
         except exceptions.ReportGenerationError as e:
-            logger.error(e)
+            logger.exception(e)
 
 
 def _write_feature_names(feature_names, output_file_root):

ms2rescore/feature_generators/ms2pip.py

@@ -53,6 +53,7 @@ def __init__(
         ms2_tolerance: float = 0.02,
         spectrum_path: Optional[str] = None,
         spectrum_id_pattern: str = "(.*)",
+        model_dir: Optional[str] = None,
         processes: 1,
         **kwargs,
     ) -> None:
@@ -71,6 +72,8 @@ def __init__(
         spectrum_id_pattern : str, optional
             Regular expression pattern to extract spectrum ID from spectrum file. Defaults to
             :py:const:`.*`.
+        model_dir
+            Directory containing MS²PIP models. Defaults to :py:const:`None` (use MS²PIP default).
         processes : int, optional
             Number of processes to use. Defaults to 1.
 
@@ -85,6 +88,7 @@ def __init__(
         self.ms2_tolerance = ms2_tolerance
         self.spectrum_path = spectrum_path
         self.spectrum_id_pattern = spectrum_id_pattern
+        self.model_dir = model_dir
         self.processes = processes
 
     @property
@@ -194,6 +198,7 @@ def add_features(self, psm_list: PSMList) -> None:
                         model=self.model,
                         ms2_tolerance=self.ms2_tolerance,
                         compute_correlations=False,
+                        model_dir=self.model_dir,
                         processes=self.processes,
                     )
                 except NoMatchingSpectraFound as e:

ms2rescore/report/charts.py

@@ -214,6 +214,14 @@ def score_scatter_plot(
         Column with index for each PSM, peptide, or protein to use for merging data frames.
 
     """
+    if not before or not after:
+        figure = go.Figure()
+        figure.add_annotation(
+            text="No data available for comparison.",
+            showarrow=False,
+        )
+        return figure
+
     # Restructure data
     merge_columns = [indexer, "mokapot score", "mokapot q-value", "mokapot PEP"]
     ce_psms_targets = pd.merge(
@@ -288,6 +296,14 @@ def fdr_plot_comparison(
         Column with index for each PSM, peptide, or protein to use for merging dataframes.
 
     """
+    if not before or not after:
+        figure = go.Figure()
+        figure.add_annotation(
+            text="No data available for comparison.",
+            showarrow=False,
+        )
+        return figure
+
     # Prepare data
     ce_psms_targets_melted = (
         pd.merge(
@@ -336,7 +352,7 @@ def fdr_plot_comparison(
 def identification_overlap(
     before: mokapot.LinearConfidence,
     after: mokapot.LinearConfidence,
-) -> go.Figure():
+) -> go.Figure:
     """
     Plot stacked bar charts of removed, retained, and gained PSMs, peptides, and proteins.
 
@@ -348,6 +364,14 @@ def identification_overlap(
         Mokapot linear confidence results after rescoring.
 
     """
+    if not before or not after:
+        figure = go.Figure()
+        figure.add_annotation(
+            text="No data available for comparison.",
+            showarrow=False,
+        )
+        return figure
+
     levels = before.levels  # ["psms", "peptides", "proteins"] if all available
     indexers = ["index", "index", "mokapot protein group"]
 

ms2rescore/report/generate.py

@@ -164,6 +164,11 @@ def _get_stats_context(confidence_before, confidence_after):
     levels = ["psms", "peptides", "proteins"]
     level_names = ["PSMs", "Peptides", "Protein groups"]
     card_colors = ["card-bg-blue", "card-bg-green", "card-bg-red"]
+
+    # Cannot report stats if confidence estimates are not present
+    if not confidence_before or not confidence_after:
+        return stats
+
     for level, level_name, card_color in zip(levels, level_names, card_colors):
         try:
             before = confidence_before.accepted[level.lower()]

ms2rescore/report/utils.py

@@ -80,10 +80,7 @@ def get_confidence_estimates(
 
         try:
             confidence[when] = lin_psm_dataset.assign_confidence()
-        except RuntimeError as e:
-            raise ReportGenerationError(
-                f"Error while assigning confidence estimates to PSMs ({when} rescoring). "
-                "Could not generate report."
-            ) from e
+        except RuntimeError:
+            confidence[when] = None
 
     return confidence["before"], confidence["after"]