Skip template mmCIFs that biopython cannot parse (#610)

colabfold/batch.py

@@ -84,6 +84,7 @@
 from colabfold.alphafold import extra_ptm, ipsae
 
 from Bio.PDB import MMCIFParser, PDBParser, MMCIF2Dict
+from Bio.PDB.PDBExceptions import PDBConstructionException
 from Bio.PDB.PDBIO import Select
 
 # logging settings
@@ -324,7 +325,11 @@ def mk_hhsearch_db(template_dir: str):
                 cif_string = f.read()
             cif_fh = StringIO(cif_string)
             parser = MMCIFParser(QUIET=True)
-            structure = parser.get_structure("none", cif_fh)
+            try:
+                structure = parser.get_structure("none", cif_fh)
+            except PDBConstructionException as e:
+                logger.warning(f"WARNING: Could not parse {cif_file} ({e}). This file is skipped as a template.")
+                continue
             models = list(structure.get_models())
             if len(models) != 1:
                 logger.warning(f"WARNING: Found {len(models)} models in {cif_file}. The first model will be used as a template.", )

tests/test_templates.py

@@ -0,0 +1,65 @@
+import logging
+
+from colabfold.batch import mk_hhsearch_db
+
+HEADER = """data_{name}
+#
+loop_
+_chem_comp.id
+_chem_comp.type
+LYS "L-peptide linking"
+#
+loop_
+_struct_asym.id
+_struct_asym.entity_id
+A 1
+#
+loop_
+_entity_poly_seq.entity_id
+_entity_poly_seq.num
+_entity_poly_seq.mon_id
+1 1 LYS
+#
+loop_
+_atom_site.group_PDB
+_atom_site.id
+_atom_site.type_symbol
+_atom_site.label_atom_id
+_atom_site.label_alt_id
+_atom_site.label_comp_id
+_atom_site.label_asym_id
+_atom_site.label_entity_id
+_atom_site.label_seq_id
+_atom_site.pdbx_PDB_ins_code
+_atom_site.Cartn_x
+_atom_site.Cartn_y
+_atom_site.Cartn_z
+_atom_site.occupancy
+_atom_site.B_iso_or_equiv
+_atom_site.auth_seq_id
+_atom_site.auth_asym_id
+_atom_site.pdbx_PDB_model_num
+ATOM 1 N N . LYS A 1 1 ? 0.000 0.000 0.000 1.00 0.00 1 A 1
+ATOM 2 C CA . LYS A 1 1 ? 1.458 0.000 0.000 1.00 0.00 1 A 1
+ATOM 3 C C . LYS A 1 1 ? 2.009 1.420 0.000 1.00 0.00 1 A 1
+ATOM 4 O O . LYS A 1 1 ? 1.251 2.390 0.000 1.00 0.00 1 A 1
+"""
+
+# like some PDB entries (e.g. 4udf, 1ejg), this one has an atom defined twice
+DUPLICATE_ATOM = "ATOM 5 C C . LYS A 1 1 ? 2.100 1.500 0.000 1.00 0.00 1 A 1\n"
+
+
+def test_mk_hhsearch_db_skips_unparsable_template(caplog, tmp_path):
+    """A template that biopython cannot parse must not abort the run (#610)"""
+    caplog.set_level(logging.WARNING)
+    tmp_path.joinpath("good.cif").write_text(HEADER.format(name="good"))
+    tmp_path.joinpath("broken.cif").write_text(
+        HEADER.format(name="broken") + DUPLICATE_ATOM
+    )
+
+    mk_hhsearch_db(str(tmp_path))
+
+    assert "broken.cif" in caplog.text
+    a3m = tmp_path.joinpath("pdb70_a3m.ffdata").read_text()
+    assert ">good_A" in a3m
+    assert ">broken_A" not in a3m