def test_organic_user_db():
db = ChemicalMetadataDB(elements=False,
main_db=os.path.join(
folder,
'chemical identifiers example user db.tsv'),
user_dbs=[])
for CAS, d in db.CAS_index.items():
assert CAS_from_any(d.CASs) == d.CASs
# Check something was loaded
assert len(db.CAS_index) > 100
# Check smiles are unique / can lookup by smiles
for smi, d in db.smiles_index.items():
if not smi:
continue
assert CAS_from_any('smiles=' + smi) == d.CASs
# Check formula is formatted right
assert all([
i.formula == serialize_formula(i.formula)
for i in db.CAS_index.values()
])
# Check CAS validity
assert all([checkCAS(i.CASs) for i in db.CAS_index.values()])
# MW checker
for i in db.CAS_index.values():
formula = serialize_formula(i.formula)
atoms = nested_formula_parser(formula, check=False)
mw_calc = molecular_weight(atoms)
assert_allclose(mw_calc, i.MW, atol=0.05)
for CAS, d in db.CAS_index.items():
assert CAS_from_any('InChI=1S/' + d.InChI) == int2CAS(CAS)
for CAS, d in db.CAS_index.items():
assert CAS_from_any('InChIKey=' + d.InChI_key) == int2CAS(CAS)
# Test the pubchem ids which aren't -1
for CAS, d in db.CAS_index.items():
if d.pubchemid != -1:
assert CAS_from_any('PubChem=' + str(d.pubchemid)) == int2CAS(CAS)
CAS_lenth = len(db.CAS_index)
assert CAS_lenth == len(db.smiles_index)
assert CAS_lenth == len(db.InChI_index)
assert CAS_lenth == len(db.InChI_key_index)
def test_TSCA_data():
tots_calc = [TSCA_data[i].sum() for i in ['UV', 'E', 'F', 'N', 'P', 'S', 'R', 'T', 'XU', 'SP', 'TP', 'Y1', 'Y2']]
tots = [16829, 271, 3, 713, 8371, 1173, 13, 151, 19035, 74, 50, 352, 9]
assert tots_calc == tots
assert TSCA_data.index.is_unique
assert TSCA_data.shape == (67635, 13)
assert all([checkCAS(int2CAS(i)) for i in TSCA_data.index])
def test_TSCA_data():
tots_calc = [TSCA_data[i].sum() for i in ['UV', 'E', 'F', 'N', 'P', 'S', 'R', 'T', 'XU', 'SP', 'TP', 'Y1', 'Y2']]
tots = [16829, 271, 3, 713, 8371, 1173, 13, 151, 19035, 74, 50, 352, 9]
assert tots_calc == tots
assert TSCA_data.index.is_unique
assert TSCA_data.shape == (67635, 13)
assert all([checkCAS(int2CAS(i)) for i in TSCA_data.index])
def test_organic_user_db():
db = ChemicalMetadataDB(elements=False,
main_db=os.path.join(folder, 'chemical identifiers example user db.tsv'),
user_dbs=[])
for CAS, d in db.CAS_index.items():
assert CAS_from_any(d.CASs) == d.CASs
# Check something was loaded
assert len(db.CAS_index) > 100
# Check smiles are unique / can lookup by smiles
for smi, d in db.smiles_index.items():
if not smi:
continue
assert CAS_from_any('smiles=' + smi) == d.CASs
# Check formula is formatted right
assert all([i.formula == serialize_formula(i.formula) for i in db.CAS_index.values()])
# Check CAS validity
assert all([checkCAS(i.CASs) for i in db.CAS_index.values()])
# MW checker
for i in db.CAS_index.values():
formula = serialize_formula(i.formula)
atoms = nested_formula_parser(formula, check=False)
mw_calc = molecular_weight(atoms)
assert_allclose(mw_calc, i.MW, atol=0.05)
for CAS, d in db.CAS_index.items():
assert CAS_from_any('InChI=1S/' + d.InChI) == int2CAS(CAS)
for CAS, d in db.CAS_index.items():
assert CAS_from_any('InChIKey=' + d.InChI_key) == int2CAS(CAS)
# Test the pubchem ids which aren't -1
for CAS, d in db.CAS_index.items():
if d.pubchemid != -1:
assert CAS_from_any('PubChem=' + str(d.pubchemid)) == int2CAS(CAS)
CAS_lenth = len(db.CAS_index)
assert CAS_lenth == len(db.smiles_index)
assert CAS_lenth == len(db.InChI_index)
assert CAS_lenth == len(db.InChI_key_index)
def test_HPV_data():
assert HPV_data.index.is_unique
assert HPV_data.shape == (5067, 0)
assert sum(list(HPV_data.index)) == 176952023632
assert all([checkCAS(int2CAS(i)) for i in HPV_data.index])
def test_NLP_data():
assert NLP_data.index.is_unique
assert NLP_data.shape == (698, 0)
assert sum(list(NLP_data.index)) == 83268755392
assert all([checkCAS(int2CAS(i)) for i in NLP_data.index])
def test_SPIN_data():
assert SPIN_data.index.is_unique
assert SPIN_data.shape == (26023, 0)
assert sum(list(SPIN_data.index)) == 1666688770043
assert all([checkCAS(int2CAS(i)) for i in SPIN_data.index])
def test_EINECS_data():
assert EINECS_data.index.is_unique
assert EINECS_data.shape == (100203, 0)
assert sum(list(EINECS_data.index)) == 4497611272838
assert all([checkCAS(int2CAS(i)) for i in EINECS_data.index])
def test_DSL_data():
assert DSL_data.index.is_unique
assert DSL_data['Registry'].sum() == 48363
assert DSL_data.shape == (73036, 1)
assert all([checkCAS(int2CAS(i)) for i in DSL_data.index])
def CASs(self):
return int2CAS(self.CAS)
def CASs(self):
return int2CAS(self.CAS)
def test_HPV_data():
assert HPV_data.index.is_unique
assert HPV_data.shape == (5067, 0)
assert sum(list(HPV_data.index)) == 176952023632
assert all([checkCAS(int2CAS(i)) for i in HPV_data.index])
def test_NLP_data():
assert NLP_data.index.is_unique
assert NLP_data.shape == (698, 0)
assert sum(list(NLP_data.index)) == 83268755392
assert all([checkCAS(int2CAS(i)) for i in NLP_data.index])
def test_SPIN_data():
assert SPIN_data.index.is_unique
assert SPIN_data.shape == (26023, 0)
assert sum(list(SPIN_data.index)) == 1666688770043
assert all([checkCAS(int2CAS(i)) for i in SPIN_data.index])
def test_EINECS_data():
assert EINECS_data.index.is_unique
assert EINECS_data.shape == (100203, 0)
assert sum(list(EINECS_data.index)) == 4497611272838
assert all([checkCAS(int2CAS(i)) for i in EINECS_data.index])
def test_DSL_data():
assert DSL_data.index.is_unique
assert DSL_data['Registry'].sum() == 48363
assert DSL_data.shape == (73036, 1)
assert all([checkCAS(int2CAS(i)) for i in DSL_data.index])
