Merge pull request #140 from PixelgenTechnologies/fix/jaccard-index-s…

…etting-negative-marker-values-to-true explicitly check marker value expressions to be greater than zero bef…
PixelgenTechnologies · May 17, 2024 · 5fb97ff · 5fb97ff
2 parents 28e0e10 + 9a8a42b
commit 5fb97ff
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 4 deletions.
diff --git a/src/pixelator/analysis/colocalization/statistics.py b/src/pixelator/analysis/colocalization/statistics.py
@@ -101,7 +101,7 @@ def jaccard(df: RegionByCountsDataFrame) -> MarkerColocalizationResults:
     :return: MarkerColocalizationResults with Jaccard indexes
     """
     jaccard_matrix = pd.DataFrame(
-        1 - pairwise_distances(df.T.to_numpy(dtype=bool), metric="jaccard"),
+        1 - pairwise_distances((df.T > 0).to_numpy(dtype=bool), metric="jaccard"),
         index=df.columns.copy(),
         columns=df.columns.copy(),
     )

diff --git a/tests/analysis/colocalization/test_colocalization.py b/tests/analysis/colocalization/test_colocalization.py
@@ -185,8 +185,8 @@ def test_colocalization_scores_ratediff(
                 "pearson_z": np.nan,
                 "pearson_p_value": np.nan,
                 "pearson_p_value_adjusted": np.nan,
-                "jaccard": 1.0,
-                "jaccard_mean": 1.0,
+                "jaccard": 0.0,
+                "jaccard_mean": 0.0,
                 "jaccard_stdev": 0.0,
                 "jaccard_z": np.nan,
                 "jaccard_p_value": np.nan,

diff --git a/tests/analysis/colocalization/test_statistics.py b/tests/analysis/colocalization/test_statistics.py
@@ -72,7 +72,7 @@ def test_pearson_no_variation():
 
 def test_jaccard():
     df = pd.DataFrame(
-        [[0, 0, 3, 4], [2, 3, 0, 6], [0, 5, 3, 8]],
+        [[0, -1, 3, 4], [2, 3, 0, 6], [0, 5, 3, 8]],
         columns=["marker1", "marker2", "marker3", "marker4"],
     )
     results = jaccard(df)