Merge pull request #27 from mattwthompson/cleanup

Remove some unused functions, attributes, logic

.github/workflows/ci.yaml

@@ -34,7 +34,7 @@ jobs:
     - name: Run tests
       run: |
         pytest -v -n logical --durations=10 \
-          --cov=ibstore/ --cov-report=xml \
+          --cov=ibstore/ --cov-report=xml --cov-config=setup.cfg \
           ibstore/
 
     - name: CodeCov

ibstore/_forcefields.py

@@ -2,33 +2,6 @@
 from openff.toolkit import Molecule
 
 
-def _get_force_field_type(
-    force_field_path: str,
-) -> str:
-    if force_field_path.endswith(".offxml"):
-        return "SMIRNOFF"
-    elif force_field_path.endswith(".xml"):
-        return "OPENMM"
-    else:
-        raise NotImplementedError(f"Force field type {force_field_path} not supported.")
-
-
-def _get_openmm_system(
-    molecule: Molecule,
-    force_field_path: str,
-) -> openmm.System:
-    force_field_type = _get_force_field_type(force_field_path)
-
-    if force_field_type == "SMIRNOFF":
-        return _smirnoff(molecule, force_field_path)
-
-    elif force_field_type == "OPENMM":
-        return _gaff(molecule, force_field_path)
-
-    else:
-        raise NotImplementedError(f"force field type {force_field_type} not supported.")
-
-
 def _smirnoff(molecule: Molecule, force_field_path: str) -> openmm.System:
     from openff.toolkit import ForceField
 

ibstore/_minimize.py

@@ -68,37 +68,21 @@ def _lazy_load_force_field(force_field_name: str) -> ForceField:
 def _minimize_blob(
     input: dict[str, dict[str, Union[str, numpy.ndarray]]],
     force_field: str,
-    prune_isomorphs: bool,
     n_processes: int = 2,
     chunksize=32,
 ) -> dict[str, list["MinimizationResult"]]:
     inputs = list()
 
-    for inchi_key in input:
-        for row in input[inchi_key]:
-            if prune_isomorphs:
-                # This behavior is always useless and probably bad as there is
-                # no reason to use InCHI when mapped SMILES is known. See #7
-                are_isomorphic, _ = Molecule.are_isomorphic(
-                    Molecule.from_inchi(inchi_key, allow_undefined_stereo=True),
-                    Molecule.from_mapped_smiles(
-                        row["mapped_smiles"],
-                        allow_undefined_stereo=True,
-                    ),
-                )
-
-                if not are_isomorphic:
-                    continue
-
-            inputs.append(
-                MinimizationInput(
-                    inchi_key=inchi_key,
-                    qcarchive_id=row["qcarchive_id"],
-                    force_field=force_field,
-                    mapped_smiles=row["mapped_smiles"],
-                    coordinates=row["coordinates"],
-                ),
-            )
+    inputs = [
+        MinimizationInput(
+            qcarchive_id=row["qcarchive_id"],
+            force_field=force_field,
+            mapped_smiles=row["mapped_smiles"],
+            coordinates=row["coordinates"],
+        )
+        for inchi_key in input
+        for row in input[inchi_key]
+    ]
 
     with Pool(processes=n_processes) as pool:
         yield from tqdm(
@@ -113,7 +97,6 @@ def _minimize_blob(
 
 
 class MinimizationInput(ImmutableModel):
-    inchi_key: str = Field(..., description="The InChI key of the molecule")
     qcarchive_id: str = Field(
         ...,
         description="The ID of the molecule in the QCArchive",
@@ -133,7 +116,6 @@ class MinimizationInput(ImmutableModel):
 
 
 class MinimizationResult(ImmutableModel):
-    inchi_key: str = Field(..., description="The InChI key of the molecule")
     qcarchive_id: str
     force_field: str
     mapped_smiles: str
@@ -144,7 +126,6 @@ class MinimizationResult(ImmutableModel):
 def _run_openmm(
     input: MinimizationInput,
 ) -> MinimizationResult:
-    inchi_key: str = input.inchi_key
     qcarchive_id: str = input.qcarchive_id
     positions: numpy.ndarray = input.coordinates
 
@@ -204,7 +185,6 @@ def _run_openmm(
     openmm.LocalEnergyMinimizer.minimize(context, 5.0e-9, 1500)
 
     return MinimizationResult(
-        inchi_key=inchi_key,
         qcarchive_id=qcarchive_id,
         force_field=input.force_field,
         mapped_smiles=input.mapped_smiles,

ibstore/_store.py

@@ -477,15 +477,14 @@ def from_cached_result_collection(
     def optimize_mm(
         self,
         force_field: str,
-        prune_isomorphs: bool = False,
         n_processes: int = 2,
         chunksize=32,
     ):
         from ibstore._minimize import _minimize_blob
 
         inchi_keys = self.get_inchi_keys()
 
-        _data = defaultdict(list)
+        inchi_key_qm_conformer_mapping = defaultdict(list)
 
         for inchi_key in inchi_keys:
             molecule_id = self.get_molecule_id_by_inchi_key(inchi_key)
@@ -509,17 +508,16 @@ def optimize_mm(
                         qcarchive_id=qm_conformer["qcarchive_id"],
                         force_field=force_field,
                     ):
-                        _data[inchi_key].append(qm_conformer)
+                        inchi_key_qm_conformer_mapping[inchi_key].append(qm_conformer)
                     else:
                         pass
 
-        if len(_data) == 0:
+        if len(inchi_key_qm_conformer_mapping) == 0:
             return
 
         _minimized_blob = _minimize_blob(
-            input=_data,
+            input=inchi_key_qm_conformer_mapping,
             force_field=force_field,
-            prune_isomorphs=prune_isomorphs,
             n_processes=n_processes,
             chunksize=chunksize,
         )
@@ -546,7 +544,6 @@ def optimize_mm(
             for result in _minimized_blob:
                 if result.qcarchive_id in seen:
                     continue
-                inchi_key = result.inchi_key
                 molecule_id = inchi_to_id[inchi_key]
                 record = MMConformerRecord(
                     molecule_id=molecule_id,

ibstore/_tests/unit_tests/test_minimize.py

@@ -30,7 +30,6 @@ def basic_input(force_field="openff-1.0.0") -> MinimizationInput:
     ethane.generate_conformers(n_conformers=1)
 
     return MinimizationInput(
-        inchi_key=ethane.to_inchikey(),
         qcarchive_id="test",
         force_field=force_field,
         mapped_smiles=ethane.to_smiles(mapped=True),
@@ -41,7 +40,6 @@ def basic_input(force_field="openff-1.0.0") -> MinimizationInput:
 @pytest.fixture()
 def perturbed_input(perturbed_ethane) -> MinimizationInput:
     return MinimizationInput(
-        inchi_key=perturbed_ethane.to_inchikey(),
         qcarchive_id="test",
         force_field="openff-1.0.0",
         mapped_smiles=perturbed_ethane.to_smiles(mapped=True),
@@ -57,7 +55,6 @@ def test_minimization_basic(perturbed_input):
     result = _run_openmm(perturbed_input)
 
     for attr in (
-        "inchi_key",
         "qcarchive_id",
         "mapped_smiles",
     ):
@@ -85,7 +82,6 @@ def test_different_force_fields_different_results():
 def test_plugin_loadable(ethane):
     _run_openmm(
         MinimizationInput(
-            inchi_key=ethane.to_inchikey(),
             qcarchive_id="test",
             force_field="de-force-1.0.1",
             mapped_smiles=ethane.to_smiles(mapped=True),
@@ -108,7 +104,6 @@ def test_finds_local_force_field(ethane, tmp_path):
 
     _run_openmm(
         MinimizationInput(
-            inchi_key=ethane.to_inchikey(),
             qcarchive_id="test",
             force_field=(tmp_path / "fOOOO.offxml").as_posix(),
             mapped_smiles=ethane.to_smiles(mapped=True),
@@ -131,7 +126,6 @@ def return_no_paths():
 
     _run_openmm(
         MinimizationInput(
-            inchi_key=ethane.to_inchikey(),
             qcarchive_id="test",
             force_field="openff-1.0.0",
             mapped_smiles=ethane.to_smiles(mapped=True),

setup.cfg

@@ -26,8 +26,9 @@ exclude =
 
 [coverage:run]
 omit =
-    */_version.py
-    */_tests/*
+    ibstore/_tests/*
+    versioneer.py
+    setup.py
     ibstore/_version.py
 
 [coverage:report]