def test_filter_by_name(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
loaded_sdf: IndigoRecordMolecule,
resource_loader,
):
mol = indigo_fixture.loadMoleculeFromFile(
resource_loader("molecules/composition1.mol"))
elastic_repository_molecule.index_record(
IndigoRecordMolecule(indigo_object=mol))
time.sleep(1)
result = elastic_repository_molecule.filter(name="Composition1")
for item in result:
assert item.name == "Composition1"
result = elastic_repository_molecule.filter(
similarity=TanimotoSimilarityMatch(
IndigoRecordMolecule(indigo_object=mol), 0.1))
i = 0
for _ in result:
i += 1
assert i == 10
result = elastic_repository_molecule.filter(
similarity=TanimotoSimilarityMatch(
IndigoRecordMolecule(indigo_object=mol), 0.1),
name="Composition1",
)
for item in result:
assert item.name == "Composition1"
def test_empty_create(indigo_fixture):
mol = indigo_fixture.createMolecule()
with pytest.raises(Exception):
IndigoRecordMolecule(indigo_object=mol)
def err_handler(instance: object, err_: BaseException) -> None:
assert isinstance(instance.record_id, str)
assert isinstance(err_, ValueError)
IndigoRecordMolecule(indigo_object=mol, error_handler=err_handler)
IndigoRecordMolecule(indigo_object=mol, skip_errors=True)
def search(data):
indigo = Indigo()
compound = indigo.loadMolecule(data)
indigo_record = IndigoRecordMolecule(indigo_object=compound)
elasticRepository = elastic_repository_molecule()
return elasticRepository.filter(exact=indigo_record)
def iterate_file(
file: Path,
iterator: str = None,
error_handler: Optional[Callable[[object, BaseException], None]] = None,
) -> Generator[IndigoRecordMolecule, None, None]:
"""
:param file:
:param iterator: supported iterators sdf, smiles, smi, cml.
If iterator is not set, trying to determine
iterator by file extension
:type iterator: str
:param error_handler: lambda for catching exceptions
:type error_handler: Optional[Callable[[object, BaseException], None]]
:return:
"""
iterators = {
"sdf": "iterateSDFile",
"smiles": "iterateSmilesFile",
"smi": "iterateSmilesFile",
"cml": "iterateCMLFile",
}
if not iterator:
iterator = file.suffix[1:]
iterator_fn = iterators.get(iterator)
if not iterator_fn:
raise AttributeError(f"Unsupported iterator {iterator}")
indigo_object: IndigoObject
for indigo_object in getattr(Indigo(), iterator_fn)(str(file)):
yield IndigoRecordMolecule(
indigo_object=indigo_object, error_handler=error_handler
)
def push(data):
indigo = Indigo()
compound = indigo.loadMolecule(data)
indigo_record = IndigoRecordMolecule(indigo_object=compound)
elasticRepository = elastic_repository_molecule()
elasticRepository.index_record(record=indigo_record)
def test_similaririty_matches_reactions(
elastic_repository_reaction: ElasticRepository,
loaded_rxns,
resource_loader,
indigo_fixture,
) -> None:
reaction = indigo_fixture.loadReactionFromFile(
resource_loader("reactions/rheadb/50353.rxn"))
reaction_rec = IndigoRecordMolecule(indigo_object=reaction)
for found_reaction in elastic_repository_reaction.filter(
similarity=TanimotoSimilarityMatch(reaction_rec, 0.99)):
assert (as_iob(
found_reaction,
indigo_fixture).countReactants() == reaction.countReactants())
for found_reaction in elastic_repository_reaction.filter(
similarity=EuclidSimilarityMatch(reaction_rec, 0.99)):
assert (as_iob(
found_reaction,
indigo_fixture).countReactants() == reaction.countReactants())
for found_reaction in elastic_repository_reaction.filter(
similarity=TverskySimilarityMatch(reaction_rec, 0.99)):
assert (as_iob(
found_reaction,
indigo_fixture).countReactants() == reaction.countReactants())
for found_reaction in elastic_repository_reaction.filter(
exact=reaction_rec):
assert (as_iob(
found_reaction,
indigo_fixture).countReactants() == reaction.countReactants())
def get_record_by_index(
response: Dict,
index: str) -> Union[IndigoRecordMolecule, IndigoRecordReaction]:
if index == IndexName.BINGO_MOLECULE.value:
return IndigoRecordMolecule(elastic_response=response)
if index == IndexName.BINGO_REACTION.value:
return IndigoRecordReaction(elastic_response=response)
raise AttributeError(f"Unknown index {index}")
def load_molecule(
file_: Union[str, Path], session: Indigo
) -> IndigoRecordMolecule:
"""
Helper for loading molecules from file into IndigoRecordMolecule object
"""
molecule = session.loadMoleculeFromFile(file_)
return IndigoRecordMolecule(indigo_object=molecule)
def test_exact_match(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
loaded_sdf: IndigoRecordMolecule,
):
result = elastic_repository_molecule.filter(exact=loaded_sdf)
assert (loaded_sdf.as_indigo_object(indigo_fixture).canonicalSmiles() ==
next(result).as_indigo_object(indigo_fixture).canonicalSmiles())
def test_substructure_search(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
loaded_sdf: IndigoRecordMolecule,
):
result = elastic_repository_molecule.filter(substructure=loaded_sdf)
for item in result:
assert (item.as_indigo_object(indigo_fixture).canonicalSmiles() ==
loaded_sdf.as_indigo_object(indigo_fixture).canonicalSmiles())
def test_search_empty_fingerprint(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
resource_loader,
):
for smile in ["[H][H]", "[H][F]"]:
rec = IndigoRecordMolecule(
indigo_object=indigo_fixture.loadMolecule(smile), skip_errors=True)
elastic_repository_molecule.index_record(rec)
time.sleep(5)
result = elastic_repository_molecule.filter(exact=IndigoRecordMolecule(
indigo_object=indigo_fixture.loadMolecule("[H][H]"),
skip_errors=True,
))
assert ("[H][H]" == next(result).as_indigo_object(
indigo_fixture).canonicalSmiles())
with pytest.raises(StopIteration):
next(result).as_indigo_object(indigo_fixture).canonicalSmiles()
def test_similarity_matches(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
loaded_sdf: IndigoRecordMolecule,
):
for sim_alg in [
TanimotoSimilarityMatch(loaded_sdf, 0.9),
EuclidSimilarityMatch(loaded_sdf, 0.9),
TverskySimilarityMatch(loaded_sdf, 0.9, 0.5, 0.5),
]:
result = elastic_repository_molecule.filter(similarity=sim_alg)
assert (loaded_sdf.as_indigo_object(indigo_fixture).canonicalSmiles(
) == next(result).as_indigo_object(indigo_fixture).canonicalSmiles())
def test_wildcard_search(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
loaded_sdf: IndigoRecordMolecule,
resource_loader,
):
mol = indigo_fixture.loadMoleculeFromFile(
resource_loader("molecules/composition1.mol"))
elastic_repository_molecule.index_record(
IndigoRecordMolecule(indigo_object=mol))
time.sleep(1)
result = elastic_repository_molecule.filter(name=WildcardQuery("Comp*"))
for item in result:
assert item.name == "Composition1"
def test_custom_fields(
elastic_repository_molecule: ElasticRepository,
indigo_fixture: Indigo,
loaded_sdf: IndigoRecordMolecule,
resource_loader,
):
mol = indigo_fixture.loadMoleculeFromFile(
resource_loader("molecules/composition1.mol"))
rec = IndigoRecordMolecule(
indigo_object=mol,
PUBCHEM_IUPAC_INCHIKEY="RDHQFKQIGNGIED-UHFFFAOYSA-N")
elastic_repository_molecule.index_record(rec)
time.sleep(1)
result = elastic_repository_molecule.filter(
PUBCHEM_IUPAC_INCHIKEY="RDHQFKQIGNGIED-UHFFFAOYSA-N")
for item in result:
assert item.PUBCHEM_IUPAC_INCHIKEY == "RDHQFKQIGNGIED-UHFFFAOYSA-N"
def test_create_with_name(indigo_fixture, resource_loader):
mol = indigo_fixture.loadMoleculeFromFile(
resource_loader("molecules/composition1.mol"))
indigo_record = IndigoRecordMolecule(indigo_object=mol)
assert indigo_record.name == "Composition1"
def test_create_without_fingerprint(indigo_fixture):
mol = indigo_fixture.loadMolecule("[H][H]")
indigo_record = IndigoRecordMolecule(indigo_object=mol, skip_errors=True)
assert len(indigo_record.sim_fingerprint) == 0
assert len(indigo_record.sub_fingerprint) == 0
def test_create(indigo_fixture):
mol = indigo_fixture.loadMolecule("N1(CC)C2=C(C(=NC=N2)N)N=C1")
indigo_record = IndigoRecordMolecule(indigo_object=mol)
assert len(indigo_record.sim_fingerprint) == 70
assert len(indigo_record.sub_fingerprint) == 644
assert len(indigo_record.cmf) == 140