fix: empty score matrix breaks MatchMS

metatlas/plots/dill2plots.py

@@ -30,6 +30,7 @@
 import pandas as pd
 import dill
 import numpy as np
+import numpy.typing as npt
 import json
 import matplotlib.pyplot as plt
 
@@ -62,6 +63,7 @@
 from io import StringIO
 
 from matchms.similarity import CosineHungarian
+from matchms.typing import SpectrumType
 from matchms import Spectrum
 
 logger = logging.getLogger(__name__)
@@ -2242,6 +2244,42 @@ def create_nonmatched_msms_hits(msms_data: pd.DataFrame, inchi_key: str) -> pd.D
 
     return inchi_msms_hits
 
+
+def score_matrix(cos: Type[CosineHungarian], references: List[SpectrumType], queries: List[SpectrumType]) -> npt.NDArray:
+    """
+    Calculate matrix of scores and matching ion counts using MatchMS spectrum objects.
+
+    This is a replacement for (and is derived from) the native MatchMS BaseSimilarity.matrix method.
+    Source for is here: https://github.com/matchms/matchms/blob/master/matchms/similarity/BaseSimilarity.py
+
+    This additional function is necessary to fix numpy errors that arise when the score matrix is filled with all 0 values.
+    """
+
+    score_datatype = [("score", "float"), ("matches", "int")]
+
+    n_rows = len(references)
+    n_cols = len(queries)
+    idx_row = []
+    idx_col = []
+    scores = []
+    for i_ref, reference in enumerate(references[:n_rows]):
+        for i_query, query in enumerate(queries[:n_cols]):
+            score = cos.pair(reference, query)
+
+            idx_row.append(i_ref)
+            idx_col.append(i_query)
+            scores.append(score)
+
+    idx_row = np.array(idx_row)
+    idx_col = np.array(idx_col)
+    scores_data = np.array(scores, dtype=score_datatype)
+
+    scores_array = np.zeros(shape=(n_rows, n_cols), dtype=score_datatype)
+    scores_array[idx_row, idx_col] = scores_data.reshape(-1)
+
+    return scores_array
+
+
 def get_hits_per_compound(cos: Type[CosineHungarian], inchi_key: str,
                           msms_data: pd.DataFrame, msms_refs: pd.DataFrame) -> pd.DataFrame:
     """
@@ -2260,8 +2298,7 @@ def get_hits_per_compound(cos: Type[CosineHungarian], inchi_key: str,
 
     filtered_msms_data = msms_data[msms_data['inchi_key']==inchi_key].reset_index(drop=True).drop(columns=['inchi_key', 'precursor_mz']).copy()
 
-    scores_matches = cos.matrix(filtered_msms_data.matchms_spectrum.tolist(),
-                                filtered_msms_refs.matchms_spectrum.tolist())
+    scores_matches = score_matrix(cos, filtered_msms_data.matchms_spectrum.tolist(), filtered_msms_refs.matchms_spectrum.tolist())
 
     inchi_msms_hits = pd.merge(filtered_msms_data, filtered_msms_refs.drop(columns=['name', 'adduct']), how='cross')
     inchi_msms_hits['score'] = scores_matches['score'].flatten()