Merge pull request #1176 from haddocking/check_combination_chains

refactored check for chain combinations

Author: mgiulini

src/haddock/libs/libcns.py

@@ -13,6 +13,7 @@
 from haddock.libs.libfunc import false, true
 from haddock.libs.libmath import RandomNumberGenerator
 from haddock.libs.libontology import PDBFile
+from haddock.libs.libpdb import check_combination_chains
 from haddock.libs.libutil import transform_to_list
 
 
@@ -335,25 +336,14 @@ def prepare_cns_input(
     # prepare chain/seg IDs
     segid_str = ""
     if native_segid:
-        chainid_list: list[str] = []
         if isinstance(input_element, (list, tuple)):
-            for pdb in input_element:
-
-                segids, chains = libpdb.identify_chainseg(pdb.rel_path, sort=False)
-
-                chainsegs = sorted(list(set(segids) | set(chains)))
-                # check if any of chainsegs is already in chainid_list
-                if not identifier.endswith("scoring"):
-                    if any(chainseg in chainid_list for chainseg in chainsegs):
-                        raise ValueError(
-                            f"Chain/seg IDs are not unique for pdbs {input_element}."
-                        )
-                chainid_list.extend(chainsegs)
+            chainid_list = check_combination_chains(input_element)
 
             for i, _chainseg in enumerate(chainid_list, start=1):
                 segid_str += write_eval_line(f"prot_segid_{i}", _chainseg)
 
         else:
+            chainid_list: list[str] = []
             segids, chains = libpdb.identify_chainseg(
                 input_element.rel_path, sort=False
             )

src/haddock/libs/libpdb.py

@@ -17,7 +17,7 @@
     Optional,
     Union,
     )
-from haddock.libs.libio import working_directory
+from haddock.libs.libio import working_directory, PDBFile
 from haddock.libs.libutil import get_result_or_same_in_list, sort_numbered_paths
 
 
@@ -296,3 +296,18 @@ def read_RECORD_section(
 
 read_chainids = partial(read_RECORD_section, section_slice=slc_chainid, func=list)  # noqa: E501
 read_segids = partial(read_RECORD_section, section_slice=slc_segid, func=list)
+
+
+def check_combination_chains(combination: list[PDBFile]) -> list[str]:
+    """Check if chain IDs are unique for each pdb in combination."""
+    chainid_list: list[str] = []
+    for pdb in combination:
+        segids, chains = identify_chainseg(pdb.rel_path, sort=False)
+        chainsegs = sorted(list(set(segids) | set(chains)))
+        # check if any of chainsegs is already in chainid_list
+        if any(chainseg in chainid_list for chainseg in chainsegs):
+            raise ValueError(
+                f"Chain/seg IDs are not unique for pdbs {combination}."
+            )
+        chainid_list.extend(chainsegs)
+    return chainid_list

src/haddock/modules/sampling/rigidbody/__init__.py

@@ -38,6 +38,7 @@
 from haddock.libs.libcns import prepare_cns_input
 from haddock.libs.libontology import PDBFile
 from haddock.libs.libparallel import GenericTask, Scheduler
+from haddock.libs.libpdb import check_combination_chains
 from haddock.libs.libsubprocess import CNSJob
 from haddock.modules import get_engine
 from haddock.modules.base_cns_module import BaseCNSModule
@@ -132,6 +133,7 @@ def prepare_cns_input_parallel(
         _l = []
         idx = 1
         for combination in models_to_dock:
+            check_combination_chains(combination)
             for _ in range(sampling_factor):
                 ambig_fname = (
                     ambig_fnames[idx - 1]

tests/test_libpdb.py

@@ -1,8 +1,12 @@
 """Test lib PDB."""
+from pathlib import Path
 import pytest
 
+
 from haddock.libs import libpdb
+from haddock.libs.libio import PDBFile
 
+from . import golden_data
 
 chainC = [
     'ATOM      3  CA  ARG C   4      37.080  43.455  -3.421  1.00  0.00      C    C  ',  # noqa: E501
@@ -31,3 +35,28 @@ def test_read_chain_ids(lines, expected):
 def test_read_seg_ids(lines, expected):
     result = libpdb.read_segids(lines)
     assert result == expected
+
+
+@pytest.fixture(name="wrong_rigid_molecules")
+def fixture_wrong_rigidbody_molecules():
+    """fixture for wrong rigidbody input molecules."""
+    receptor = PDBFile(Path(golden_data, "protprot_complex_1.pdb"))
+    ligand = PDBFile(Path(golden_data, "protprot_complex_2.pdb"))
+    return [receptor, ligand]
+
+@pytest.fixture(name="good_rigid_molecules")
+def fixture_good_rigidbody_molecules():
+    """fixture for good rigidbody input molecules."""
+    receptor = PDBFile(Path(golden_data, "e2aP_1F3G_haddock.pdb"))
+    ligand = PDBFile(Path(golden_data, "hpr_ensemble_1_haddock.pdb"))
+    return [receptor, ligand]
+
+def test_check_combination_chains(good_rigid_molecules, wrong_rigid_molecules):
+    """Test check_combination_chains."""
+    exp_chains = ["A", "B"]
+    obs_chains = libpdb.check_combination_chains(good_rigid_molecules)
+    assert obs_chains == exp_chains
+    # when input molecules share chains there should be a ValueError
+    with pytest.raises(ValueError):
+        libpdb.check_combination_chains(wrong_rigid_molecules)
+