def get_smiles(base_path, molecules, extra=False):
"""
Returns SMILES strings for a set of molecules
:param base_path: Directory to begin search in.
:param molecules: List of strings representing molecule IDs.
:param extra: If True (default False), give molecule ID with SMILES
:return: list of SMILES strings
"""
smiles = []
if base_path is not None:
for mol in molecules:
path = join(base_path, mol)
molfile = join(path, "{}.mol".format(mol))
obmol = BabelMolAdaptor.from_file(molfile, file_format="mol")
smi = obmol.pybel_mol.write("smi").replace("\t\n", "")
if extra:
smiles.append((smi, mol))
else:
smiles.append(smi)
return smiles
else:
raise ValueError("No path given.")
def run(self, site_property=None):
"""
Write the input file to the scratch directory, run packmol and return
the packed molecule to the current working directory.
Args:
site_property (str): if set then the specified site property
for the the final packed molecule will be restored.
Returns:
Molecule object
"""
with tempfile.TemporaryDirectory() as scratch_dir:
self._write_input(input_dir=scratch_dir)
with open(os.path.join(scratch_dir, self.input_file),
"r") as packmol_input:
with Popen(self.packmol_bin,
stdin=packmol_input,
stdout=PIPE,
stderr=PIPE) as p:
(stdout, stderr) = p.communicate()
output_file = os.path.join(self.control_params["output"])
if os.path.isfile(output_file):
packed_mol = BabelMolAdaptor.from_file(
output_file, self.control_params["filetype"])
packed_mol = packed_mol.pymatgen_mol
print("packed molecule written to {}".format(
self.control_params["output"]))
if site_property:
packed_mol = self.restore_site_properties(
site_property=site_property, filename=output_file)
return packed_mol
raise RuntimeError("Packmol execution failed. %s\n%s" %
(stdout, stderr))
def test_from_file_return_all_molecules(self):
adaptors = BabelMolAdaptor.from_file(
os.path.join(PymatgenTest.TEST_FILES_DIR, "multiple_frame_xyz.xyz"),
"xyz",
return_all_molecules=True,
)
self.assertEqual(len(adaptors), 302)
def restore_site_properties(self, site_property="ff_map", filename=None):
"""
Restore the site properties for the final packed molecule.
Args:
site_property (str):
filename (str): path to the final packed molecule.
Returns:
Molecule
"""
# only for pdb
if not self.control_params["filetype"] == "pdb":
raise
filename = filename or self.control_params["output"]
bma = BabelMolAdaptor.from_file(filename, "pdb")
pbm = pb.Molecule(bma._obmol)
assert len(pbm.residues) == sum([x["number"] for x in self.param_list])
packed_mol = self.convert_obatoms_to_molecule(
pbm.residues[0].atoms, residue_name=pbm.residues[0].name,
site_property=site_property)
for resid in pbm.residues[1:]:
mol = self.convert_obatoms_to_molecule(
resid.atoms, residue_name=resid.name, site_property=site_property)
for site in mol:
packed_mol.append(site.species_and_occu, site.coords,
properties=site.properties)
return packed_mol
def run(self, copy_to_current_on_exit=False, site_property=None):
"""
Write the input file to the scratch directory, run packmol and return
the packed molecule.
Args:
copy_to_current_on_exit (bool): Whether or not to copy the packmol
input/output files from the scratch directory to the current
directory.
site_property (str): if set then the specified site property
for the the final packed molecule will be restored.
Returns:
Molecule object
"""
scratch = tempfile.gettempdir()
with ScratchDir(scratch, copy_to_current_on_exit=copy_to_current_on_exit) as scratch_dir:
self._write_input(input_dir=scratch_dir)
packmol_input = open(os.path.join(scratch_dir, self.input_file), 'r')
p = Popen(self.packmol_bin, stdin=packmol_input, stdout=PIPE, stderr=PIPE)
(stdout, stderr) = p.communicate()
output_file = os.path.join(scratch_dir, self.control_params["output"])
if os.path.isfile(output_file):
packed_mol = BabelMolAdaptor.from_file(output_file,
self.control_params["filetype"])
packed_mol = packed_mol.pymatgen_mol
print("packed molecule written to {}".format(
self.control_params["output"]))
if site_property:
packed_mol = self.restore_site_properties(site_property=site_property, filename=output_file)
return packed_mol
else:
print("Packmol execution failed")
print(stdout, stderr)
return None
def restore_site_properties(self, site_property="ff_map", filename=None):
"""
Restore the site properties for the final packed molecule.
Args:
site_property (str):
filename (str): path to the final packed molecule.
Returns:
Molecule
"""
# only for pdb
if not self.control_params["filetype"] == "pdb":
raise ValueError()
filename = filename or self.control_params["output"]
bma = BabelMolAdaptor.from_file(filename, "pdb")
pbm = pb.Molecule(bma._obmol)
assert len(pbm.residues) == sum(x["number"] for x in self.param_list)
packed_mol = self.convert_obatoms_to_molecule(
pbm.residues[0].atoms,
residue_name=pbm.residues[0].name,
site_property=site_property,
)
for resid in pbm.residues[1:]:
mol = self.convert_obatoms_to_molecule(resid.atoms, residue_name=resid.name, site_property=site_property)
for site in mol:
packed_mol.append(site.species, site.coords, properties=site.properties)
return packed_mol
def run(self, copy_to_current_on_exit=False, site_property=None):
"""
Write the input file to the scratch directory, run packmol and return
the packed molecule.
Args:
copy_to_current_on_exit (bool): Whether or not to copy the packmol
input/output files from the scratch directory to the current
directory.
site_property (str): if set then the specified site property
for the the final packed molecule will be restored.
Returns:
Molecule object
"""
scratch = tempfile.gettempdir()
with ScratchDir(scratch, copy_to_current_on_exit=copy_to_current_on_exit) as scratch_dir:
self._write_input(input_dir=scratch_dir)
packmol_input = open(os.path.join(scratch_dir, self.input_file), 'r')
p = Popen(self.packmol_bin, stdin=packmol_input, stdout=PIPE, stderr=PIPE)
(stdout, stderr) = p.communicate()
output_file = os.path.join(scratch_dir, self.control_params["output"])
if os.path.isfile(output_file):
packed_mol = BabelMolAdaptor.from_file(output_file,
self.control_params["filetype"])
packed_mol = packed_mol.pymatgen_mol
print("packed molecule written to {}".format(
self.control_params["output"]))
if site_property:
packed_mol = self.restore_site_properties(site_property=site_property, filename=output_file)
return packed_mol
else:
print("Packmol execution failed")
print(stdout, stderr)
return None
def run(self, copy_to_current_on_exit=False):
"""
Write the input file to the scratch directory, run packmol and return
the packed molecule.
Args:
copy_to_current_on_exit (bool): Whether or not to copy the packmol
input/output files from the scratch directory to the current
directory.
Returns:
Molecule object
"""
scratch = tempfile.gettempdir()
with ScratchDir(scratch, copy_to_current_on_exit=copy_to_current_on_exit) as scratch_dir:
self._write_input(input_dir=scratch_dir)
packmol_bin = ['packmol']
packmol_input = open(os.path.join(scratch_dir, self.input_file), 'r')
p = Popen(packmol_bin, stdin=packmol_input, stdout=PIPE, stderr=PIPE)
p.wait()
(stdout, stderr) = p.communicate()
output_file = os.path.join(scratch_dir, self.control_params["output"])
if os.path.isfile(output_file):
packed_mol = BabelMolAdaptor.from_file(output_file)
print("packed molecule written to {}".format(
self.control_params["output"]))
return packed_mol.pymatgen_mol
else:
print("Packmol execution failed")
print(stdout, stderr)
return None
def read_mol(filename):
"""
Reads a molecule based on file extension. For example, anything ending in
a "xyz" is assumed to be a XYZ file. Supported formats include xyz,
gaussian input (gjf|g03|g09|com|inp), Gaussian output (.out|and
pymatgen's JSON serialized molecules. Using openbabel,
many more extensions are supported but requires openbabel to be installed.
Args:
filename (str): A filename to read from.
Returns:
A Molecule object.
"""
fname = os.path.basename(filename)
if fnmatch(fname.lower(), "*.xyz*"):
return XYZ.from_file(filename).molecule
elif any([fnmatch(fname.lower(), "*.{}*".format(r)) for r in ["gjf", "g03", "g09", "com", "inp"]]):
return GaussianInput.from_file(filename).molecule
elif any([fnmatch(fname.lower(), "*.{}*".format(r)) for r in ["out", "lis", "log"]]):
return GaussianOutput(filename).final_structure
elif fnmatch(fname, "*.json*") or fnmatch(fname, "*.mson*"):
with zopen(filename) as f:
s = json.load(f, cls=MontyDecoder)
if type(s) != Molecule:
raise IOError("File does not contain a valid serialized " "molecule")
return s
else:
m = re.search("\.(pdb|mol|mdl|sdf|sd|ml2|sy2|mol2|cml|mrv)", filename.lower())
if m:
return BabelMolAdaptor.from_file(filename, m.group(1)).pymatgen_mol
raise ValueError("Unrecognized file extension!")
def run(self):
"""
Runs packmol
Returns:
Molecule object
"""
scratch = tempfile.gettempdir()
with ScratchDir(scratch, copy_to_current_on_exit=True) as d:
with open(os.path.join(d, self.input_file), 'wt',
encoding="utf-8") as inp:
for k, v in six.iteritems(self.control_params):
inp.write('{} {}\n'.format(k, self._format_param_val(v)))
for idx, mol in enumerate(self.mols):
a = BabelMolAdaptor(mol)
pm = pb.Molecule(a.openbabel_mol)
pm.write(self.control_params["filetype"],
filename=os.path.join(d, '{}.{}'.format(idx,
self.control_params[
"filetype"])).encode(
"ascii"),
overwrite=True)
inp.write("\n")
inp.write(
"structure {}.{}\n".format(os.path.join(d, str(idx)),
self.control_params[
"filetype"]))
for k, v in six.iteritems(self.param_list[idx]):
inp.write(
' {} {}\n'.format(k, self._format_param_val(v)))
inp.write('end structure\n')
proc = Popen(['packmol'],
stdin=open(os.path.join(d, self.input_file), 'r'),
stdout=PIPE)
(stdout, stderr) = proc.communicate()
output_file = os.path.join(d, self.control_params["output"])
if os.path.isfile(output_file):
packed_mol = BabelMolAdaptor.from_file(output_file)
print("packed molecule written to {}".format(
self.control_params["output"]))
return packed_mol.pymatgen_mol
else:
print("Packmol execution failed")
print(stdout, stderr)
return None
def get_molecule(molfile):
"""
Create pymatgen Molecule object from molecule data file.
In addition to parsing the input, this function also performs a conformer
search to get a reasonable starting structure.
:param molfile: Absolute path to structure file (.mol, .sdf, etc.)
:return: Molecule.
"""
obmol = BabelMolAdaptor.from_file(molfile, file_format="mol")
# OBMolecule does not contain pymatgen Molecule information
# So, we need to wrap the obmol in a BabelMolAdapter and extract
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
return obmol.pymatgen_mol
def read_mol(filename):
"""
Reads a molecule based on file extension. For example, anything ending in
a "xyz" is assumed to be a XYZ file. Supported formats include xyz,
gaussian input (gjf|g03|g09|com|inp), Gaussian output (.out|and
pymatgen's JSON serialized molecules. Using openbabel,
many more extensions are supported but requires openbabel to be installed.
Args:
filename (str): A filename to read from.
Returns:
A Molecule object.
"""
fname = os.path.basename(filename)
if fnmatch(fname.lower(), "*.xyz*"):
return XYZ.from_file(filename).molecule
elif any([
fnmatch(fname.lower(), "*.{}*".format(r))
for r in ["gjf", "g03", "g09", "com", "inp"]
]):
return GaussianInput.from_file(filename).molecule
elif any([
fnmatch(fname.lower(), "*.{}*".format(r))
for r in ["out", "lis", "log"]
]):
return GaussianOutput(filename).final_structure
elif fnmatch(fname, "*.json*") or fnmatch(fname, "*.mson*"):
with zopen(filename) as f:
s = json.load(f, cls=MontyDecoder)
if type(s) != Molecule:
raise IOError("File does not contain a valid serialized "
"molecule")
return s
else:
m = re.search("\.(pdb|mol|mdl|sdf|sd|ml2|sy2|mol2|cml|mrv)",
filename.lower())
if m:
return BabelMolAdaptor.from_file(filename, m.group(1)).pymatgen_mol
raise ValueError("Unrecognized file extension!")
def write_pdb(self, mol, filename, name=None, num=None):
"""
dump the molecule into pdb file with custom residue name and number.
"""
# ugly hack to get around the openbabel issues with inconsistent
# residue labelling.
scratch = tempfile.gettempdir()
with ScratchDir(scratch, copy_to_current_on_exit=False) as _:
mol.to(fmt="pdb", filename="tmp.pdb")
bma = BabelMolAdaptor.from_file("tmp.pdb", "pdb")
num = num or 1
name = name or "ml{}".format(num)
# bma = BabelMolAdaptor(mol)
pbm = pb.Molecule(bma._obmol)
for i, x in enumerate(pbm.residues):
x.OBResidue.SetName(name)
x.OBResidue.SetNum(num)
pbm.write(format="pdb", filename=filename, overwrite=True)
def write_pdb(self, mol, filename, name=None, num=None):
"""
dump the molecule into pdb file with custom residue name and number.
"""
# ugly hack to get around the openbabel issues with inconsistent
# residue labelling.
scratch = tempfile.gettempdir()
with ScratchDir(scratch, copy_to_current_on_exit=False) as _:
mol.to(fmt="pdb", filename="tmp.pdb")
bma = BabelMolAdaptor.from_file("tmp.pdb", "pdb")
num = num or 1
name = name or "ml{}".format(num)
# bma = BabelMolAdaptor(mol)
pbm = pb.Molecule(bma._obmol)
for i, x in enumerate(pbm.residues):
x.OBResidue.SetName(name)
x.OBResidue.SetNum(num)
pbm.write(format="pdb", filename=filename, overwrite=True)
def test_from_file(self):
adaptor = BabelMolAdaptor.from_file(
os.path.join(PymatgenTest.TEST_FILES_DIR, "molecules/Ethane_e.pdb"), "pdb"
)
mol = adaptor.pymatgen_mol
self.assertEqual(mol.formula, "H6 C2")
def main():
def gcd(a, b):
if b == 0:
return a
else:
return gcd(b, a % b)
def lcm(a, b):
return a * b / gcd(a, b)
import argparse
parser = argparse.ArgumentParser(
description="Place salt around a molecule")
parser.add_argument("-m", "--molecule", dest="molecule", type=str,
required=True,
help="the file name of molecule")
parser.add_argument("-l", "--ligand", dest="fragments", type=str,
nargs='+',
required=True,
help="the list of fragment file names to to be placed around the molecule")
parser.add_argument("-n", "--nums_fragments", dest="nums_fragments",
type=int, nargs='+',
required=True,
help="the number of each fragment, the order must be the same with FRAGMENTS")
parser.add_argument("-c", "--charge", dest="charge", type=int,
required=True,
help="total charge of the system")
parser.add_argument("-t", "--taboo_tolerance", dest="taboo_tolerance",
type=float,
default=1.0,
help="The radius to taboo a solution (in Angstrom)")
parser.add_argument("-r", "--ratio_taboo_particles",
dest="ratio_taboo_particles", type=float,
default=0.5,
help="ratio of particle within the tolerance to consider taboo current solution")
parser.add_argument("-o", "--outputfile", dest="outputfile", type=str,
required=True,
help="the file name of the aligned conformer")
parser.add_argument("-i", "--iterations", dest="iterations", type=int,
default=600,
help="maximum number of evaluations")
parser.add_argument("-s", "--size", dest="size", type=int, default=15,
help="population size")
parser.add_argument("-k", "--num_neighbours", dest="num_neighbours",
type=int, default=2,
help="number of neighbours")
parser.add_argument("--force_ordered_fragment",
dest="force_ordered_fragment", action="store_true",
help="set this option to keep the fragment of the same in the order of input along the X-axis")
parser.add_argument("--topology", dest="topology",
choices=["ring", "star"], type=str, default="ring",
help="the topology of the PSO information network")
parser.add_argument("--initial_guess", dest="initial_guess",
choices=["breadth", "center", "volume"],
default="breadth",
help="where should particles should be initially put")
parser.add_argument("--bound_setter", dest="bound_setter",
choices=["chain", "volume"], default="chain",
help="method to set the bound conditions of PSO")
parser.add_argument("--always_write_best", dest="always_write_best",
action="store_true",
help="enable this option to output the best structure at every iteration")
parser.add_argument("--random_seed", dest="random_seed", default=None,
type=int,
help="random seed for PSO, an integer is expected")
parser.add_argument("--max_generations_each_conformer",
dest="max_generations_each_conformer", default=100,
type=int,
help="maximum generations for each conformer")
parser.add_argument("-e", "--evaluator", dest="evaluator", type=str,
default="hardsphere",
choices=["hardsphere", "sqm"], help="Energy Evaluator")
options = parser.parse_args()
if options.evaluator == 'hardsphere':
qcout_molecule = QcOutput(options.molecule)
qcout_cation = QcOutput(options.cation)
qcout_anion = QcOutput(options.anion)
total_charge_cation = qcout_cation.data[0]["molecules"][-1].charge
total_charge_anion = qcout_anion.data[0]["molecules"][-1].charge
total_charge_mol = qcout_molecule.data[0]["molecules"][-1].charge
num_lcm = lcm(total_charge_cation, -total_charge_anion)
num_cation = num_lcm / total_charge_cation
num_anion = num_lcm / -total_charge_anion
pymatgen_mol_molecule = qcout_molecule.data[0]["molecules"][-1]
pymatgen_mol_cation = qcout_cation.data[0]["molecules"][-1]
pymatgen_mol_anion = qcout_anion.data[0]["molecules"][-1]
# noinspection PyProtectedMember
molecule = BabelMolAdaptor(pymatgen_mol_molecule)._obmol
# noinspection PyProtectedMember
obmol_cation = BabelMolAdaptor(pymatgen_mol_cation)._obmol
# noinspection PyProtectedMember
obmol_anion = BabelMolAdaptor(pymatgen_mol_anion)._obmol
energy_evaluator = HardSphereElectrostaticEnergyEvaluator.from_qchem_output(
qcout_molecule, qcout_cation, qcout_anion)
fragments = [obmol_cation, obmol_anion]
else:
# noinspection PyProtectedMember
molecule = BabelMolAdaptor.from_file(options.molecule,
os.path.splitext(
options.molecule)[1][
1:])._obmol
fragments = []
for frag_file in options.fragments:
file_format = os.path.splitext(frag_file)[1][1:]
# noinspection PyProtectedMember
fragments.append(
BabelMolAdaptor.from_file(frag_file, file_format)._obmol)
energy_evaluator = SemiEmpricalQuatumMechanicalEnergyEvaluator(
molecule, fragments, options.nums_fragments,
total_charge=options.charge,
taboo_tolerance_ang=options.taboo_tolerance,
force_order_fragment=options.force_ordered_fragment,
bound_setter=options.bound_setter)
if len(fragments) != len(options.nums_fragments):
raise ValueError(
"you must specify the duplicated count for every fragment")
placer = IonPlacer(molecule=molecule, fragments=fragments,
nums_fragments=options.nums_fragments,
energy_evaluator=energy_evaluator,
taboo_tolerance_ang=options.taboo_tolerance,
taboo_tolerance_particle_ratio=options.ratio_taboo_particles,
topology=options.topology,
initial_guess=options.initial_guess,
bound_setter=options.bound_setter,
always_write_best=options.always_write_best,
random_seed=options.random_seed,
max_generations_each_conformer=options.max_generations_each_conformer)
energy_evaluator.arranger = placer
placer.place(max_evaluations=options.iterations,
pop_size=options.size,
neighborhood_size=options.num_neighbours)
print('It took {:.1f} seconds to place the salt'.format(placer
.playing_time))
def test_from_file(self):
adaptor = BabelMolAdaptor.from_file(
os.path.join(test_dir, "Ethane_e.pdb"), "pdb")
mol = adaptor.pymatgen_mol
self.assertEqual(mol.formula, "H6 C2")
def test_from_file_return_all_molecules(self):
adaptors = BabelMolAdaptor.from_file(os.path.join(
test_dir, "multiple_frame_xyz.xyz"),
"xyz",
return_all_molecules=True)
self.assertEqual(len(adaptors), 302)
def __init__(self, molecule, optimize=False):
"""
Instantiation method for FunctionalGroupExtractor.
:param molecule: Either a filename, a pymatgen.core.structure.Molecule
object, or a pymatgen.analysis.graphs.MoleculeGraph object.
:param optimize: Default False. If True, then the input molecule will be
modified, adding Hydrogens, performing a simple conformer search,
etc.
"""
self.molgraph = None
if isinstance(molecule, str):
try:
if optimize:
obmol = BabelMolAdaptor.from_file(molecule,
file_format="mol")
# OBMolecule does not contain pymatgen Molecule information
# So, we need to wrap the obmol in a BabelMolAdapter
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
self.molecule = obmol.pymatgen_mol
else:
self.molecule = Molecule.from_file(molecule)
except OSError:
raise ValueError("Input must be a valid molecule file, a "
"Molecule object, or a MoleculeGraph object.")
elif isinstance(molecule, Molecule):
if optimize:
obmol = BabelMolAdaptor(molecule)
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
self.molecule = obmol.pymatgen_mol
else:
self.molecule = molecule
elif isinstance(molecule, MoleculeGraph):
if optimize:
obmol = BabelMolAdaptor(molecule.molecule)
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
self.molecule = obmol.pymatgen_mol
else:
self.molecule = molecule.molecule
self.molgraph = molecule
else:
raise ValueError("Input to FunctionalGroupExtractor must be"
"str, Molecule, or MoleculeGraph.")
if self.molgraph is None:
self.molgraph = MoleculeGraph.with_local_env_strategy(self.molecule,
OpenBabelNN(),
reorder=False,
extend_structure=False)
# Assign a specie and coordinates to each node in the graph,
# corresponding to the Site in the Molecule object
self.molgraph.set_node_attributes()
self.species = nx.get_node_attributes(self.molgraph.graph, "specie")
def __init__(self, molecule, optimize=False):
"""
Instantiation method for FunctionalGroupExtractor.
:param molecule: Either a filename, a pymatgen.core.structure.Molecule
object, or a pymatgen.analysis.graphs.MoleculeGraph object.
:param optimize: Default False. If True, then the input molecule will be
modified, adding Hydrogens, performing a simple conformer search,
etc.
"""
self.molgraph = None
if isinstance(molecule, str):
try:
if optimize:
obmol = BabelMolAdaptor.from_file(molecule,
file_format="mol")
# OBMolecule does not contain pymatgen Molecule information
# So, we need to wrap the obmol in a BabelMolAdapter
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
self.molecule = obmol.pymatgen_mol
else:
self.molecule = Molecule.from_file(molecule)
except OSError:
raise ValueError("Input must be a valid molecule file, a "
"Molecule object, or a MoleculeGraph object.")
elif isinstance(molecule, Molecule):
if optimize:
obmol = BabelMolAdaptor(molecule)
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
self.molecule = obmol.pymatgen_mol
else:
self.molecule = molecule
elif isinstance(molecule, MoleculeGraph):
if optimize:
obmol = BabelMolAdaptor(molecule.molecule)
obmol.add_hydrogen()
obmol.make3d()
obmol.localopt()
self.molecule = obmol.pymatgen_mol
else:
self.molecule = molecule.molecule
self.molgraph = molecule
else:
raise ValueError("Input to FunctionalGroupExtractor must be"
"str, Molecule, or MoleculeGraph.")
if self.molgraph is None:
self.molgraph = MoleculeGraph.with_local_env_strategy(
self.molecule,
OpenBabelNN(),
reorder=False,
extend_structure=False)
# Assign a specie and coordinates to each node in the graph,
# corresponding to the Site in the Molecule object
self.molgraph.set_node_attributes()
self.species = nx.get_node_attributes(self.molgraph.graph, "specie")
def test_from_file(self):
adaptor = BabelMolAdaptor.from_file(
os.path.join(test_dir, "Ethane_e.pdb"), "pdb")
mol = adaptor.pymatgen_mol
self.assertEqual(mol.formula, "H6 C2")