fix: SCHWED-5783 distinguish between 0 lDDT and no match

modules/mol/alg/pymod/ligand_scoring.py

@@ -460,7 +460,7 @@ class LigandScorer:
                     #                " - setting to Infinity" % str(err))
                     #     bb_rmsd = float("inf")
                     lddt_pli_full_matrix[target_i, model_i] = {
-                        "lddt_pli": 0,
+                        "lddt_pli": np.nan,
                         "lddt_pli_n_contacts": None,
                         "rmsd": rmsd,
                         # "symmetry_number": i,
@@ -496,7 +496,7 @@ class LigandScorer:
                         # Save results?
                         best_lddt = lddt_pli_full_matrix[
                             target_i, model_i]["lddt_pli"]
-                        if global_lddt > best_lddt:
+                        if global_lddt > best_lddt or np.isnan(best_lddt):
                             lddt_pli_full_matrix[target_i, model_i].update({
                                 "lddt_pli": global_lddt,
                                 "lddt_pli_n_contacts": lddt_tot,
@@ -515,8 +515,8 @@ class LigandScorer:
         mat1 = np.copy(mat1)
         mat2 = np.copy(mat2)
         assignments = []
-        min_mat1 = mat1.min()
-        while min_mat1 < np.inf:
+        min_mat1 = LigandScorer._nanmin_nowarn(mat1)
+        while not np.isnan(min_mat1):
             best_mat1 = np.argwhere(mat1 == min_mat1)
             # Multiple "best" - use mat2 to disambiguate
             if len(best_mat1) > 1:
@@ -530,18 +530,25 @@ class LigandScorer:
                 max_i_trg, max_i_mdl = best_mat1[0]
 
             # Disable row and column
-            mat1[max_i_trg, :] = np.inf
-            mat1[:, max_i_mdl] = np.inf
-            mat2[max_i_trg, :] = np.inf
-            mat2[:, max_i_mdl] = np.inf
+            mat1[max_i_trg, :] = np.nan
+            mat1[:, max_i_mdl] = np.nan
+            mat2[max_i_trg, :] = np.nan
+            mat2[:, max_i_mdl] = np.nan
 
             # Save
             assignments.append((max_i_trg, max_i_mdl))
 
             # Recompute min
-            min_mat1 = mat1.min()
+            min_mat1 = LigandScorer._nanmin_nowarn(mat1)
         return assignments
 
+    @staticmethod
+    def _nanmin_nowarn(array):
+        """Compute np.nanmin but ignore the RuntimeWarning."""
+        with warnings.catch_warnings():  # RuntimeWarning: All-NaN slice encountered
+            warnings.simplefilter("ignore")
+            return np.nanmin(array)
+
     @staticmethod
     def _reverse_lddt(lddt):
         """Reverse lDDT means turning it from a number between 0 and 1 to a
@@ -610,7 +617,7 @@ class LigandScorer:
 
         Target ligands are in rows, model ligands in columns.
 
-        Infinite values indicate that no RMSD could be computed (i.e. different
+        NaN values indicate that no RMSD could be computed (i.e. different
         ligands).
 
         :rtype: :class:`~numpy.ndarray`
@@ -620,7 +627,7 @@ class LigandScorer:
         if self._rmsd_matrix is None:
             # convert
             shape = self._rmsd_full_matrix.shape
-            self._rmsd_matrix = np.full(shape, np.inf)
+            self._rmsd_matrix = np.full(shape, np.nan)
             for i, j in np.ndindex(shape):
                 if self._rmsd_full_matrix[i, j] is not None:
                     self._rmsd_matrix[i, j] = self._rmsd_full_matrix[
@@ -633,7 +640,7 @@ class LigandScorer:
 
         Target ligands are in rows, model ligands in columns.
 
-        A value of 0 indicate that no lDDT-PLI could be computed (i.e. different
+        NaN values indicate that no lDDT-PLI could be computed (i.e. different
         ligands).
 
         :rtype: :class:`~numpy.ndarray`
@@ -643,7 +650,7 @@ class LigandScorer:
         if self._lddt_pli_matrix is None:
             # convert
             shape = self._lddt_pli_full_matrix.shape
-            self._lddt_pli_matrix = np.zeros(shape)
+            self._lddt_pli_matrix = np.full(shape, np.nan)
             for i, j in np.ndindex(shape):
                 if self._lddt_pli_full_matrix[i, j] is not None:
                     self._lddt_pli_matrix[i, j] = self._lddt_pli_full_matrix[

modules/mol/alg/tests/test_ligand_scoring.py

@@ -251,23 +251,23 @@ class TestLigandScoring(unittest.TestCase):
         # Check RMSD
         assert sc.rmsd_matrix.shape == (7, 1)
         np.testing.assert_almost_equal(sc.rmsd_matrix, np.array(
-            [[np.inf],
+            [[np.nan],
             [0.04244993],
-            [np.inf],
-            [np.inf],
-            [np.inf],
+            [np.nan],
+            [np.nan],
+            [np.nan],
             [0.29399303],
-            [np.inf]]), decimal=5)
+            [np.nan]]), decimal=5)
 
         # Check lDDT-PLI
         self.assertEqual(sc.lddt_pli_matrix.shape, (7, 1))
-        self.assertEqual(sc.lddt_pli_matrix[0, 0], 0)
+        self.assertTrue(np.isnan(sc.lddt_pli_matrix[0, 0]))
         self.assertAlmostEqual(sc.lddt_pli_matrix[1, 0], 0.99843, 5)
-        self.assertEqual(sc.lddt_pli_matrix[2, 0], 0)
-        self.assertEqual(sc.lddt_pli_matrix[3, 0], 0)
-        self.assertEqual(sc.lddt_pli_matrix[4, 0], 0)
+        self.assertTrue(np.isnan(sc.lddt_pli_matrix[2, 0]))
+        self.assertTrue(np.isnan(sc.lddt_pli_matrix[3, 0]))
+        self.assertTrue(np.isnan(sc.lddt_pli_matrix[4, 0]))
         self.assertAlmostEqual(sc.lddt_pli_matrix[5, 0], 1.0)
-        self.assertEqual(sc.lddt_pli_matrix[6, 0], 0)
+        self.assertTrue(np.isnan(sc.lddt_pli_matrix[6, 0]))
 
     def test_check_resnames(self):
         """Test check_resname argument works