def test_SubSearchAccuracy():
db_mock = utils.setupMockDB()
write.WriteFromSDF(db_mock.molecules, 'data/test_data/first_200.props.sdf')
substructure.AddPatternFingerprints(db_mock.molecules)
db_python = utils.setupPythonDB('data/test_data/first_200.props.sdf')
for i in range(200):
moldoc = db_python[i]
pattern = Chem.Mol(moldoc['rdmol'])
results_python = utils.SubSearchPython(pattern, db_python)
results = substructure.SubSearch(pattern, db_mock.molecules)
assert sorted(results_python) == sorted(results)
def test_zeroThreshold():
"""
If we search the database with a Tanimoto threshold of zero, we should get
the entire contents of the database back.
"""
db_python = utils.setupPythonDB('data/test_data/first_200.props.sdf')
db_mongo = utils.setupMockDB()
write.WriteFromSDF(db_mongo.molecules, 'data/test_data/first_200.props.sdf')
similarity.AddMorganFingerprints(db_mongo.molecules, db_mongo.mfp_counts)
mol = Chem.Mol(db_python[0]['rdmol'])
assert 200 == len(utils.similaritySearchPython(mol, db_python, 0))
assert 200 == len(similarity.SimSearchAggregate(mol, db_mongo.molecules, db_mongo.mfp_counts, 0))
assert 200 == len(similarity.SimSearch(mol, db_mongo.molecules, db_mongo.mfp_counts, 0))
def test_similarityAccuracy():
"""
Tests for basic accuracy against a brute-force constructed Python 'database'
at thresholds 0.2, 0.4, 0.6, 0.8, and 1. This test is implemented using MongoMock.
"""
db_python = utils.setupPythonDB('data/test_data/first_200.props.sdf')
db_mongo = utils.setupMockDB()
write.WriteFromSDF(db_mongo.molecules, 'data/test_data/first_200.props.sdf')
similarity.AddMorganFingerprints(db_mongo.molecules, db_mongo.mfp_counts)
thresholds = [0.2, 0.4, 0.6, 0.8, 1]
for t in thresholds:
for i in range(200):
mol = Chem.Mol(db_python[i]['rdmol'])
search_python = utils.similaritySearchPython(mol, db_python, t)
search_mongo = similarity.SimSearch(mol, db_mongo.molecules, db_mongo.mfp_counts, t)
assert sorted(search_python) == sorted(search_mongo)
def test_similarityProgression():
"""
Tests that decreasing similarity thresholds return increasing result lists.
This test is implemented using MongoMock.
"""
db_python = utils.setupPythonDB('data/test_data/first_200.props.sdf')
db_mongo = utils.setupMockDB()
write.WriteFromSDF(db_mongo.molecules, 'data/test_data/first_200.props.sdf')
similarity.AddMorganFingerprints(db_mongo.molecules, db_mongo.mfp_counts)
thresholds = [1, 0.8, 0.6, 0.4, 0.2]
for i in range(200):
mol = Chem.Mol(db_python[i]['rdmol'])
last = []
for t in thresholds:
search_mongo = similarity.SimSearch(mol, db_mongo.molecules, db_mongo.mfp_counts, t)
assert len(search_mongo) >= len(last)
assert (all(l in search_mongo for l in last))
last = search_mongo
def test_addPatternFingerprints():
db = utils.setupMockDB()
write.WriteFromSDF(db.molecules, 'data/test_data/first_200.props.sdf')
substructure.AddPatternFingerprints(db.molecules)
counter = 0
assert db.molecules.count_documents({"fingerprints.pattern_fp": {"$exists": True}}) == 200