Add tests for disulfides

devtools/conda-envs/docs_env.yaml

@@ -5,7 +5,6 @@ dependencies:
   # Base depends
   - python>=3.11
   - pip
-  - pyxdg
 
   # Documentation
   - myst-parser
@@ -25,6 +24,7 @@ dependencies:
   - typer>=0.14
   - mdanalysis
   - rdkit
+  - pyxdg
 
   #  Pip-only installs
   - pip:

devtools/conda-envs/test_env.yaml

@@ -2,15 +2,25 @@ name: pablo-test
 channels:
   - conda-forge
 dependencies:
-  # Base depends
+  # Base depends (runtime and imports)
   - pip
   - python>=3.11
+  - openff-toolkit-base>=0.14.7
+  - rdkit
+  - openmm
   - pyxdg
 
   # Lints
   - basedpyright
   - ruff
 
+  # Testing
+  - pytest
+  - pytest-xdist
+  - pytest-socket
+  # - pytest-cov
+  # - codecov
+
   # Examples
   - jupyterlab>=4
   - ipywidgets>=8
@@ -22,24 +32,13 @@ dependencies:
   - black
   - isort
   - snakeviz
-
-  # Testing
-  - pytest
-  # - pytest-cov
-  - pytest-xdist
-  - pytest-socket
-  # - codecov
-
-  # chemistry
-  - openff-toolkit-base>=0.14.7
-  - rdkit
   - nglview>=3.0.6,<3.1.0
   - pdbfixer
   - mdanalysis
   - openff-interchange
   - ambertools
+  - numpy
 
   - pip:
-      # - codecov
       - git+https://github.com/openforcefield/openff-toolkit.git@better_typing
       - git+https://github.com/openforcefield/openff-units.git@better_typing

docs/tests.md

@@ -15,6 +15,10 @@ Pablo has a (developing) PDB test suite.
     - ACE/NME caps, `unique_molecules`
 *   - 193L_prepared.pdb
     - disulfides, all 20 canonical amino acids, charged terminii, charged canonical AA side chains
+*   - 2hi7_prepared.pdb
+    - disulfide bond between two different protein chains
+*   - 2zuq_prepared.pdb
+    - disulfide bond between two different protein chains, with additional chains between and after the crosslinked chains
 :::
 
 <script>

openff/pablo/_pdb.py

@@ -1,4 +1,5 @@
 import itertools
+import warnings
 from collections.abc import Iterable, Mapping, MutableSequence
 from io import TextIOBase
 from os import PathLike
@@ -344,6 +345,8 @@ def topology_from_pdb(
     n = len(topology_pdb_indices)
     positions = np.stack([data.x[:n], data.y[:n], data.z[:n]], axis=-1) * unit.angstrom
     topology.set_positions(positions[topology_pdb_indices])
+    if topology_pdb_indices != list(range(n)):
+        warnings.warn("Atoms in topology are not in same order as those in PDB file")
 
     if set_stereochemistry_from_3d:
         for molecule in topology.molecules:
@@ -362,7 +365,7 @@ def topology_from_pdb(
 
 def _check_all_conects(topology: Topology, data: PdbData):
     all_bonds: set[tuple[int, int]] = {
-        sort_tuple((topology.atom_index(bond.atom1), topology.atom_index(bond.atom2)))
+        sort_tuple((bond.atom1.metadata["pdb_index"], bond.atom2.metadata["pdb_index"]))  # type:ignore
         for bond in topology.bonds
     }
 
@@ -501,12 +504,12 @@ def _add_to_molecule(
             return this_molecule
 
         for other_molecule in molecules:
-            other_mol_pdb_idx_to_mol_atom_idx = other_molecule.properties[
-                "pdb_idx_to_mol_atom_idx"
-            ]
+            other_mol_pdb_idx_to_mol_atom_idx: dict[int, int] = (
+                other_molecule.properties["pdb_idx_to_mol_atom_idx"]
+            )
             assert isinstance(
-                dict,
                 other_mol_pdb_idx_to_mol_atom_idx,
+                dict,
             ), "This property should have already been set by Pablo"
 
             if other_idx in other_mol_pdb_idx_to_mol_atom_idx:

openff/pablo/_tests/conftest.py

@@ -1,34 +1,38 @@
 from pathlib import Path
 
 import pytest
-from pkg_resources import resource_filename
 
 from openff.pablo._pdb_data import PdbData, ResidueMatch
+from openff.pablo._tests.utils import get_test_data_path
 from openff.pablo.chem import DISULFIDE_BOND, PEPTIDE_BOND
 from openff.pablo.residue import AtomDefinition, BondDefinition, ResidueDefinition
 
 
 @pytest.fixture(
     params=[
-        "data/5ap1_prepared.pdb",
-        "data/193l_prepared.pdb",
-        "data/3cu9_vicinal_disulfide.pdb",
-        "data/e2_7nel.pdb",
+        "5ap1_prepared.pdb",
+        "prepared_pdbs/193l_prepared.pdb",
+        "prepared_pdbs/2zuq_prepared.pdb",
+        "prepared_pdbs/2hi7_prepared.pdb",
+        "3cu9_vicinal_disulfide.pdb",
+        "e2_7nel.pdb",
     ],
 )
 def pdbfn(request: pytest.FixtureRequest) -> Path:
-    return Path(resource_filename(__name__, request.param))
+    return get_test_data_path(request.param)
 
 
 @pytest.fixture
 def hewl_data() -> PdbData:
-    return PdbData.from_file(resource_filename(__name__, "data/193l_prepared.pdb"))
+    return PdbData.from_file(
+        get_test_data_path("prepared_pdbs/193l_prepared.pdb"),
+    )
 
 
 @pytest.fixture
 def vicinal_disulfide_data() -> PdbData:
     return PdbData.from_file(
-        resource_filename(__name__, "data/3cu9_vicinal_disulfide.pdb"),
+        get_test_data_path("3cu9_vicinal_disulfide.pdb"),
     )
 
 
@@ -61,7 +65,7 @@ def cys_data(cys_pdblines: list[str]) -> PdbData:
 
 @pytest.fixture
 def e2_data() -> PdbData:
-    return PdbData.from_file(resource_filename(__name__, "data/e2_7nel.pdb"))
+    return PdbData.from_file(get_test_data_path("e2_7nel.pdb"))
 
 
 @pytest.fixture