Merge pull request #780 from haddocking/fix_prot_segid

Fix prot segid

Author: mgiulini

examples/run_tests.py

@@ -153,14 +153,6 @@ def main(examples, break_on_errors=True):
                     stderr=sys.stderr,
                     )
 
-                subprocess.run(
-                    "haddock3 docking-protein-protein-test-start-from-cp.cfg --extend-run run2",  # noqa: E501
-                    shell=True,
-                    check=break_on_errors,
-                    stdout=sys.stdout,
-                    stderr=sys.stderr,
-                    )
-
                 # test exit with extend-run
                 rmtree("run2", ignore_errors=True)
                 subprocess.run(

src/haddock/gear/prepare_run.py

@@ -1043,22 +1043,8 @@ def populate_topology_molecule_params(topoaa: ParamMap) -> None:
     """Populate topoaa `molX` subdictionaries."""
     topoaa_dft = _read_defaults("topoaa.1")
 
-    # list of possible prot_segids
-    uppers = list(string.ascii_uppercase)[::-1]
-
-    # removes from the list those prot_segids that are already defined
-    for param in topoaa:
-        if param.startswith("mol") and param[3:].isdigit():
-            with suppress(KeyError):
-                uppers.remove(topoaa[param]["prot_segid"])
-
-    # populates the prot_segids just for those that were not defined
-    # in the user configuration file. Other parameters are populated as
-    # well. `prot_segid` is the only one differing per molecule.
     for i in range(1, len(topoaa["molecules"]) + 1):
         mol = f"mol{i}"
-        if not (mol in topoaa and "prot_segid" in topoaa[mol]):
-            topoaa_dft["mol1"]["prot_segid"] = uppers.pop()
 
         topoaa[mol] = recursive_dict_update(
             topoaa_dft["mol1"],

src/haddock/libs/libcns.py

@@ -330,6 +330,12 @@ def prepare_cns_input(
                     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)
 
             for i, _chainseg in enumerate(chainid_list, start=1):

src/haddock/libs/libpdb.py

@@ -218,6 +218,11 @@ def identify_chainseg(pdb_file_path: FilePath,
                     segids.append(segid)
                 if chainid:
                     chains.append(chainid)
+                
+                if not segid and not chainid:
+                    raise ValueError(
+                        f"Could not identify chainID or segID in pdb {pdb_file_path}, line {line}"
+                        )
 
     if sort:
         segids = sorted(list(set(segids)))

src/haddock/modules/topology/topoaa/defaults.yaml

@@ -87,16 +87,6 @@ tolerance:
   group: module
   explevel: expert
 mol1:
-  prot_segid:
-    default: A
-    type: string
-    minchars: 0
-    maxchars: 4
-    title: Segment ID
-    short: Segment ID assigned to this molecule
-    long: Segment ID assigned to this molecule in CNS. Used to distinguish different molecules
-    group: molecule
-    explevel: easy
   cyclicpept:
     default: false
     type: boolean

tests/test_gear_prepare_run.py

@@ -56,8 +56,6 @@ def test_populate_topoaa_molecules():
         }
     populate_topology_molecule_params(topoaa)
     assert "mol2" in topoaa
-    assert topoaa["mol2"]["prot_segid"] == "B"
-    assert topoaa["mol1"]["prot_segid"] == "A"
     assert topoaa["mol2"]["cyclicpept"] is False
     assert topoaa["mol1"]["cyclicpept"] is True
     assert isnan(topoaa["mol2"]["hisd_1"])
@@ -74,12 +72,10 @@ def test_populate_topoaa_molecules_2():
     """Test mols are polated."""
     topoaa = {
         "molecules": ["file1.pdb", "file2.pdb"],
-        "mol2": {"cyclicpept": True, "prot_segid": "D"},
+        "mol2": {"cyclicpept": True},
         }
     populate_topology_molecule_params(topoaa)
     assert "mol1" in topoaa
-    assert topoaa["mol1"]["prot_segid"] == "A"
-    assert topoaa["mol2"]["prot_segid"] == "D"
 
     assert topoaa["mol1"]["cyclicpept"] is False
     assert topoaa["mol2"]["cyclicpept"] is True
@@ -99,27 +95,20 @@ def test_populate_topoaa_molecules_3():
     """Test mols are polated."""
     topoaa = {
         "molecules": ["file1.pdb", "file2.pdb", "file3.pdb"],
-        "mol2": {"cyclicpept": True, "prot_segid": "C"},
+        "mol2": {"cyclicpept": True},
         }
     populate_topology_molecule_params(topoaa)
     assert "mol1" in topoaa
-    assert topoaa["mol1"]["prot_segid"] == "A"
-    assert topoaa["mol2"]["prot_segid"] == "C"
-    assert topoaa["mol3"]["prot_segid"] == "B"
 
 
 def test_populate_topoaa_molecules_4():
     """Test mols are polated with prot_segid sequence."""
     topoaa = {
         "molecules": ["file1.pdb", "file2.pdb", "file3.pdb", "file4.pdb"],
-        "mol3": {"cyclicpept": True, "prot_segid": "A"},
+        "mol3": {"cyclicpept": True},
         }
     populate_topology_molecule_params(topoaa)
     assert "mol1" in topoaa
-    assert topoaa["mol1"]["prot_segid"] == "B"
-    assert topoaa["mol2"]["prot_segid"] == "C"
-    assert topoaa["mol3"]["prot_segid"] == "A"
-    assert topoaa["mol4"]["prot_segid"] == "D"
 
 
 def test_populate_mol_params():