Merge pull request #144 from compomics/peptide-num-fix

Fixes for MS²Rescore report

ms2rescore/core.py

@@ -175,8 +175,8 @@ def _fill_missing_precursor_info(psm_list, config):
         get_missing_values(psm_list, config, rt_required=rt_required, im_required=im_required)
 
     # Check if values are now present
-    for value_name in ["retention_time", "ion_mobility"]:
-        if (
+    for value_name, required in [("retention_time", rt_required), ("ion_mobility", im_required)]:
+        if required and (
             0.0 in psm_list[value_name]
             or None in psm_list[value_name]
             or np.isnan(psm_list[value_name]).any()

ms2rescore/report/charts.py

@@ -373,7 +373,7 @@ def identification_overlap(
         return figure
 
     levels = before.levels  # ["psms", "peptides", "proteins"] if all available
-    indexers = ["index", "index", "mokapot protein group"]
+    indexers = ["index", "peptide", "mokapot protein group"]
 
     overlap_data = defaultdict(dict)
     for level, indexer in zip(levels, indexers):
@@ -386,7 +386,7 @@ def identification_overlap(
         set_after = set(df_after[df_after["mokapot q-value"] <= 0.01][indexer])
 
         overlap_data["removed"][level] = -len(set_before - set_after)
-        overlap_data["retained"][level] = len(set_before | set_after)
+        overlap_data["retained"][level] = len(set_after.intersection(set_before))
         overlap_data["gained"][level] = len(set_after - set_before)
 
     colors = ["#953331", "#316395", "#319545"]

ms2rescore/report/generate.py

@@ -185,7 +185,7 @@ def _get_stats_context(confidence_before, confidence_after):
                 "item": level_name,
                 "card_color": card_color,
                 "number": after,
-                "diff": f"{after - before:+}",
+                "diff": f"({after - before:+})",
                 "percentage": f"{increase:.1f}%",
                 "is_increase": increase > 0,
                 "bar_percentage": before / after * 100 if increase > 0 else after / before * 100,

ms2rescore/rescoring_engines/mokapot.py

@@ -31,6 +31,7 @@
 from pyteomics.mass import nist_mass
 
 logger = logging.getLogger(__name__)
+logging.getLogger("numba").setLevel(logging.WARNING)
 
 
 def rescore(
@@ -89,7 +90,7 @@ def rescore(
 
     # Rescore
     logger.debug(f"Mokapot brew options: `{kwargs}`")
-    confidence_results, models = brew(lin_psm_data, **kwargs)
+    confidence_results, models = brew(lin_psm_data, rng=8, **kwargs)
 
     # Reshape confidence estimates to match PSMList
     mokapot_values_targets = (

ms2rescore/rescoring_engines/percolator.py

@@ -20,8 +20,8 @@
 import logging
 import subprocess
 from typing import Any, Dict, Optional
+from copy import deepcopy
 
-import numpy as np
 import psm_utils
 
 from ms2rescore.exceptions import MS2RescoreError
@@ -103,8 +103,15 @@ def rescore(
     # Need to be able to link back to original PSMs, so reindex spectrum IDs, but copy PSM list
     # to avoid modifying original...
     # TODO: Better approach for this?
-    psm_list_reindexed = psm_list.copy()
-    psm_list_reindexed["spectrum_id"] = np.arange(len(psm_list_reindexed))
+
+    psm_list_reindexed = deepcopy(psm_list)
+    psm_list_reindexed.set_ranks()
+    psm_list_reindexed["spectrum_id"] = [
+        f"{psm.get_usi(as_url=False)}_{psm.rank}" for psm in psm_list_reindexed
+    ]
+    spectrum_id_index = {
+        spectrum_id: index for index, spectrum_id in enumerate(psm_list_reindexed["spectrum_id"])
+    }
 
     _write_pin_file(psm_list_reindexed, pin_filepath)
 
@@ -134,10 +141,13 @@ def rescore(
         psm_list,
         percolator_kwargs["results-psms"],
         percolator_kwargs["decoy-results-psms"],
+        spectrum_id_index,
     )
 
 
-def _update_psm_scores(psm_list: psm_utils.PSMList, target_pout: str, decoy_pout: str):
+def _update_psm_scores(
+    psm_list: psm_utils.PSMList, target_pout: str, decoy_pout: str, spectrum_id_index: list
+):
     """
     Update PSM scores with Percolator results.
 
@@ -150,7 +160,9 @@ def _update_psm_scores(psm_list: psm_utils.PSMList, target_pout: str, decoy_pout
     psm_list_percolator = psm_utils.PSMList(psm_list=target_psms.psm_list + decoy_psms.psm_list)
 
     # Sort by reindexed spectrum_id so order matches original PSM list
-    psm_list_percolator[np.argsort(psm_list_percolator["spectrum_id"])]
+    psm_list_percolator = sorted(
+        psm_list_percolator, key=lambda psm: spectrum_id_index[psm["spectrum_id"]]
+    )
 
     if not len(psm_list) == len(psm_list_percolator):
         raise MS2RescoreError(