def test_failing_files():
fn = 'zzz.zzz'
with pytest.raises(IOError):
chemper_utils.get_full_path(fn)
with pytest.raises(IOError):
chemper_utils.get_data_path(fn)
def check_mol_file(file_name):
"""
Parameters
----------
file_name - str
path to a molecule file
Returns
-------
path - str
absolute path to a molecule file
raises error if file isn't available
"""
# is it a local file?
if os.path.exists(file_name):
return os.path.abspath(file_name)
path = get_data_path(os.path.join('molecules', file_name))
if not os.path.exists(path):
raise IOError(
"Molecule file (%s) was not found locally or in chemper/data/molecules"
% file_name)
return path
def mols_from_mol2(mol2_file):
"""
Parses a mol2 file into chemper molecules using RDKit
This is a hack for separating mol2 files taken from a Source Forge discussion here:
https://www.mail-archive.com/[email protected]/msg01510.html
It splits up a mol2 file into blocks and then uses RDKit to parse those blocks
Parameters
----------
mol2_file: str
relative or absolute path to a mol2 file you want to parse
accessible form the current directory
Returns
-------
mols: list of chemper Mols
list of molecules in the mol2 file as chemper molecules
"""
# TODO: check that this works with mol2 files with a single molecule
# TODO: figure out if @<TRIPOS>MOLECULE is the only delimiter acceptable in this file type
import os
if not os.path.exists(mol2_file):
from chemper.chemper_utils import get_data_path
mol_path = get_data_path(os.path.join('molecules', mol2_file))
if not os.path.exists(mol_path):
raise IOError("File '%s' not found locally or in chemper/data/molecules." % mol_file)
else:
mol2_file = mol_path
delimiter="@<TRIPOS>MOLECULE"
if mol2_file.split('.')[-1] != "mol2":
raise IOError("File '%s' is not a mol2 file" % mol2_file)
if not os.path.exists(mol2_file):
raise IOError("File '%s' not found." % mol2_file)
molecules = list()
mol2_block = list()
file_open = open(mol2_file, 'r')
for line in file_open:
if line.startswith(delimiter) and mol2_block:
rdmol = Chem.MolFromMol2Block("".join(mol2_block))
if rdmol is not None:
molecules.append(Mol(rdmol))
mol2_block = []
mol2_block.append(line)
if mol2_block:
rdmol = Chem.MolFromMol2Block("".join(mol2_block))
if rdmol is not None:
molecules.append(Mol(rdmol))
file_open.close()
return molecules
def test_valid_files(fn):
import chemper
import os
ref_path = os.path.join(chemper.__path__[0], 'data')
ref_path = os.path.join(ref_path, fn)
data_path = chemper_utils.get_data_path(fn)
assert data_path == ref_path
full_path = chemper_utils.get_full_path(fn)
assert full_path == data_path
assert full_path == ref_path
def mols_from_file(mol_file):
"""
Parses a standard molecule file into chemper molecules using OpenEye toolkits
Parameters
----------
mol_file: str
relative or full path to molecule containing the molecule file
that is accessible from the current working directory
Returns
-------
mols: list of chemper Mols
list of molecules in the mol2 file as chemper Mols
"""
import os
if not os.path.exists(mol_file):
from chemper.chemper_utils import get_data_path
mol_path = get_data_path(os.path.join('molecules', mol_file))
if not os.path.exists(mol_path):
raise IOError(
"File '%s' not found locally or in chemper/data/molecules." %
mol_file)
else:
mol_file = mol_path
molecules = list()
# make Openeye input file stream
ifs = oechem.oemolistream(mol_file)
oemol = oechem.OECreateOEGraphMol()
while oechem.OEReadMolecule(ifs, oemol):
# if an SD file, the molecule name may be in the SD tags
if oemol.GetTitle() == '':
name = oechem.OEGetSDData(oemol, 'name').strip()
oemol.SetTitle(name)
# Append to list.
molecules.append(Mol(oechem.OEMol(oemol)))
ifs.close()
return molecules
def test_matching_smirks(smirks1, smirks2, checks):
path = chemper_utils.get_data_path(
os.path.join('molecules', 'MiniDrugBank_tripos.mol2'))
mols = mol_toolkit.mols_from_mol2(path)
did_it_work = chemper_utils.check_smirks_agree(smirks1, smirks2, mols)
assert did_it_work == checks
def test_bond_exception(toolkit):
with pytest.raises(TypeError):
toolkit.Bond(None)
@pytest.mark.parametrize('toolkit', mts)
def test_mol_exception(toolkit):
with pytest.raises(TypeError):
toolkit.Mol(None)
# -------------------------------
# check molecule file parsers
# -------------------------------
mol2_abs_file = chemper_utils.get_data_path('molecules/MiniDrugBank_tripos.mol2')
mol2_rel_path = 'MiniDrugBank_tripos.mol2'
paths = [mol2_abs_file, mol2_rel_path]
# For the following functions, we will test default behavior and
# look for exceptions based on the available mol toolkit
@pytest.mark.parametrize('toolkit,path', itertools.product(mts, paths))
def test_file_parsing(toolkit,path):
mols = toolkit.mols_from_mol2(path)
assert len(mols) == 363
@pytest.mark.parametrize('path', paths)
def test_mols_specified_toolkit(path):
