def fit_with_mapper(self, mapper):
coords = [
[0.000000, 0.000000, 0.000000],
[0.000000, 0.000000, 1.089000],
[1.026719, 0.000000, -0.363000],
[-0.513360, -0.889165, -0.363000],
[-0.513360, 0.889165, -0.363000],
]
mol1 = Molecule(["C", "H", "H", "H", "H"], coords)
op = SymmOp.from_origin_axis_angle([0, 0, 0], [0.1, 0.2, 0.3], 60)
rotcoords = [op.operate(c) for c in coords]
mol2 = Molecule(["C", "H", "H", "H", "H"], rotcoords)
mm = MoleculeMatcher(mapper=mapper)
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "benzene1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "benzene2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "benzene1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "t2.xyz"))
self.assertFalse(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "c1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "c2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "t3.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "t4.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "j1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "j2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "ethene1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "ethene2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "toluene1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "toluene2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "cyclohexane1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "cyclohexane2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mol1 = Molecule.from_file(os.path.join(test_dir, "oxygen1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "oxygen2.xyz"))
self.assertTrue(mm.fit(mol1, mol2))
mm = MoleculeMatcher(tolerance=0.001, mapper=mapper)
mol1 = Molecule.from_file(os.path.join(test_dir, "t3.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "t4.xyz"))
self.assertFalse(mm.fit(mol1, mol2))
def test_to_and_from_dict(self):
mm = MoleculeMatcher(tolerance=0.5,
mapper=InchiMolAtomMapper(angle_tolerance=50.0))
d = mm.as_dict()
mm2 = MoleculeMatcher.from_dict(d)
self.assertEqual(d, mm2.as_dict())
mm = MoleculeMatcher(tolerance=0.5, mapper=IsomorphismMolAtomMapper())
d = mm.as_dict()
mm2 = MoleculeMatcher.from_dict(d)
self.assertEqual(d, mm2.as_dict())
def test_thiane_ethynyl(self):
mm = MoleculeMatcher(tolerance=0.05, mapper=InchiMolAtomMapper())
mol1 = Molecule.from_file(os.path.join(test_dir,
"thiane_ethynyl1.sdf"))
mol2 = Molecule.from_file(os.path.join(test_dir,
"thiane_ethynyl2.sdf"))
self.assertFalse(mm.fit(mol1, mol2))
def test_get_rmsd(self):
mm = MoleculeMatcher()
mol1 = BabelMolAdaptor.from_file(os.path.join(test_dir,
"t3.xyz")).pymatgen_mol
mol2 = BabelMolAdaptor.from_file(os.path.join(test_dir,
"t4.xyz")).pymatgen_mol
self.assertEqual('{0:7.3}'.format(mm.get_rmsd(mol1, mol2)), "0.00488")
def test_random_seed(self, water, ethanol):
"""
Confirm that seed = -1 generates random structures
while seed = 1 is deterministic
"""
mm = MoleculeMatcher()
# deterministic output
with tempfile.TemporaryDirectory() as scratch_dir:
pw = PackmolBoxGen(
seed=1,
inputfile="input.in",
outputfile="output.xyz",
).get_input_set(
# scratch_dir,
molecules=[
{
"name": "water",
"number": 10,
"coords": water
},
{
"name": "ethanol",
"number": 20,
"coords": ethanol
},
], )
pw.write_input(scratch_dir)
pw.run(scratch_dir)
out1 = Molecule.from_file(os.path.join(scratch_dir, "output.xyz"))
pw.run(scratch_dir)
out2 = Molecule.from_file(os.path.join(scratch_dir, "output.xyz"))
assert mm.fit(out1, out2)
# randomly generated structures
with tempfile.TemporaryDirectory() as scratch_dir:
pw = PackmolBoxGen(
seed=-1,
inputfile="input.in",
outputfile="output.xyz",
).get_input_set(molecules=[
{
"name": "water",
"number": 10,
"coords": water
},
{
"name": "ethanol",
"number": 20,
"coords": ethanol
},
], )
pw.write_input(scratch_dir)
pw.run(scratch_dir)
out1 = Molecule.from_file(os.path.join(scratch_dir, "output.xyz"))
pw.run(scratch_dir)
out2 = Molecule.from_file(os.path.join(scratch_dir, "output.xyz"))
assert not mm.fit(out1, out2)
def test_confab_conformers(self):
mol = pb.readstring("smi", "CCCC").OBMol
adaptor = BabelMolAdaptor(mol)
adaptor.make3d()
conformers = adaptor.confab_conformers()
self.assertEquals(adaptor.openbabel_mol.NumRotors(), 1)
self.assertGreaterEqual(len(conformers), 1)
if len(conformers) > 1:
self.assertNotAlmostEqual(
MoleculeMatcher().get_rmsd(conformers[0], conformers[1]), 0)
def test_group_molecules(self):
mm = MoleculeMatcher(tolerance=0.001)
with open(os.path.join(test_dir, "mol_list.txt")) as f:
filename_list = [line.strip() for line in f.readlines()]
mol_list = [read_mol(os.path.join(test_dir, f)) for f in filename_list]
mol_groups = mm.group_molecules(mol_list)
filename_groups = [[filename_list[mol_list.index(m)] for m in g]
for g in mol_groups]
with open(os.path.join(test_dir, "grouped_mol_list.txt")) as f:
grouped_text = f.read().strip()
self.assertEqual(str(filename_groups), grouped_text)
def test_group_molecules(self):
mm = MoleculeMatcher(tolerance=0.001)
filename_list = None
with open(os.path.join(test_dir, "mol_list.txt")) as f:
filename_list = [line.strip() for line in f.readlines()]
mol_list = [BabelMolAdaptor.from_file(os.path.join(test_dir, f)).pymatgen_mol\
for f in filename_list]
mol_groups = mm.group_molecules(mol_list)
filename_groups = [[filename_list[mol_list.index(m)] for m in g] for g \
in mol_groups]
grouped_text = None
with open(os.path.join(test_dir, "grouped_mol_list.txt")) as f:
grouped_text = f.read().strip()
self.assertEqual(str(filename_groups), grouped_text)
def add_if_belongs(self, cand_snl, exact_match=True):
# no need to compare if structue is different
if cand_snl.snlgroup_key != self.canonical_snl.snlgroup_key:
return False
# make sure the structure is not already in all_structures
if cand_snl.snl_id in self.all_snl_ids:
print('WARNING: add_if_belongs() has detected that you are ' \
'trying to add the same SNL id twice!')
return False
if exact_match:
mm = MoleculeMatcher(
tolerance=0.01, mapper=InchiMolAtomMapper(angle_tolerance=5.0))
if not mm.fit(cand_snl.structure, self.canonical_structure):
return False
# everything checks out, add to the group
self.all_snl_ids.append(cand_snl.snl_id)
self.updated_at = datetime.datetime.utcnow()
return True
def test_cdi_23(self):
mm = MoleculeMatcher(tolerance=0.05, mapper=InchiMolAtomMapper())
mol1 = Molecule.from_file(os.path.join(test_dir, "cdi_23_1.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "cdi_23_2.xyz"))
self.assertFalse(mm.fit(mol1, mol2))
def test_strange_inchi(self):
mm = MoleculeMatcher(tolerance=0.05, mapper=InchiMolAtomMapper())
mol1 = Molecule.from_file(os.path.join(test_dir, "k1.sdf"))
mol2 = Molecule.from_file(os.path.join(test_dir, "k2.sdf"))
self.assertTrue(mm.fit(mol1, mol2))
def test_get_rmsd(self):
mm = MoleculeMatcher()
mol1 = Molecule.from_file(os.path.join(test_dir, "t3.xyz"))
mol2 = Molecule.from_file(os.path.join(test_dir, "t4.xyz"))
self.assertEqual('{0:7.3}'.format(mm.get_rmsd(mol1, mol2)), "0.00488")
def test_thiane(self):
mm = MoleculeMatcher(tolerance=0.05, mapper=InchiMolAtomMapper())
mol1 = read_mol(os.path.join(test_dir, "thiane1.sdf"))
mol2 = read_mol(os.path.join(test_dir, "thiane2.sdf"))
self.assertFalse(mm.fit(mol1, mol2))
def __init__(self, redundancy_parameters, geometry):
'''
Makes a RedundancyGuard, and sets default parameter values if
necessary.
TODO: currently using pymatgen's structure matcher for comparing bulk
and sheet structures, both pymatgen's structure matcher and
molecule matcher for comparing wires, and only the molecule matcher
for clusters. The sheet and wire cases aren't ideal, since the
structure matcher assumes periodicity in all three dimensions, and
the molecule matcher assumes no periodicity.
Args:
redundancy parameters: a dictionary of parameters
geometry: the Geometry object
'''
# defaults
#
# lattice length tolerance, in fractional coordinates
self.default_lattice_length_tol = 0.05
# lattice angle tolerance, in degrees
self.default_lattice_angle_tol = 2
# site tolerance, in fraction of average free length per atom
self.default_site_tol = 0.1
# whether to transform to primitive cells before comparing
self.default_use_primitive_cell = True
# whether to check if structures are equal to supercells of each other
self.default_attempt_supercell = True
# RMSD tolerance for comparing clusters
self.default_rmsd_tol = 0.1
# the epa difference interval
self.default_epa_diff = 0.0
# set to defaults
if redundancy_parameters in (None, 'default'):
self.set_all_to_defaults()
# parse the parameters, and set to defaults if necessary
else:
# lattice length tolerance
if 'lattice_length_tol' not in redundancy_parameters:
self.lattice_length_tol = self.default_lattice_length_tol
elif redundancy_parameters['lattice_length_tol'] in (None,
'default'):
self.lattice_length_tol = self.default_lattice_length_tol
else:
self.lattice_length_tol = redundancy_parameters[
'lattice_length_tol']
# lattice angle tolerance
if 'lattice_angle_tol' not in redundancy_parameters:
self.lattice_angle_tol = self.default_lattice_angle_tol
elif redundancy_parameters['lattice_angle_tol'] in (None,
'default'):
self.lattice_angle_tol = self.default_lattice_angle_tol
else:
self.lattice_angle_tol = redundancy_parameters[
'lattice_angle_tol']
# site tolerance
if 'site_tol' not in redundancy_parameters:
self.site_tol = self.default_site_tol
elif redundancy_parameters['site_tol'] in (None, 'default'):
self.site_tol = self.default_site_tol
else:
self.site_tol = redundancy_parameters['site_tol']
# whether to use primitive cells
if 'use_primitive_cell' not in redundancy_parameters:
self.use_primitive_cell = self.default_use_primitive_cell
elif redundancy_parameters['use_primitive_cell'] in (None,
'default'):
self.use_primitive_cell = self.default_use_primitive_cell
else:
self.use_primitive_cell = redundancy_parameters[
'use_primitive_cell']
# whether to try matching supercells
if 'attempt_supercell' not in redundancy_parameters:
self.attempt_supercell = self.default_attempt_supercell
elif redundancy_parameters['attempt_supercell'] in (None,
'default'):
self.attempt_supercell = self.default_attempt_supercell
else:
self.attempt_supercell = redundancy_parameters[
'attempt_supercell']
# RMSD tolerance
if 'rmsd_tol' not in redundancy_parameters:
self.rmsd_tol = self.default_rmsd_tol
elif redundancy_parameters['rmsd_tol'] in (None, 'default'):
self.rmsd_tol = self.default_rmsd_tol
else:
self.rmsd_tol = redundancy_parameters['rmsd_tol']
# epa difference
if 'epa_diff' not in redundancy_parameters:
self.epa_diff = self.default_epa_diff
elif redundancy_parameters['epa_diff'] in (None, 'default'):
self.epa_diff = self.default_epa_diff
else:
self.epa_diff = redundancy_parameters['epa_diff']
# make the StructureMatcher object
#
# the first False is to prevent the matcher from scaling the volumes,
# and the second False is to prevent subset matching
self.structure_matcher = StructureMatcher(
self.lattice_length_tol, self.site_tol, self.lattice_angle_tol,
self.use_primitive_cell, False, self.attempt_supercell, False,
ElementComparator())
# make the MoleculeMatcher object
if geometry.shape == 'cluster' or geometry.shape == 'wire':
iso_mol_atom_mapper = IsomorphismMolAtomMapper()
self.molecule_matcher = MoleculeMatcher(self.rmsd_tol,
iso_mol_atom_mapper)
ob.obErrorLog.SetOutputLevel(0) # to suppress openbabel warnings
