def test_ring_breaking_detection():
"""
Test the detection of ring-breaking transformations.
"""
from perses.rjmc.topology_proposal import SmallMoleculeSetProposalEngine, AtomMapper
from openmoltools.openeye import iupac_to_oemol, generate_conformers
molecule1 = iupac_to_oemol("naphthalene")
molecule2 = iupac_to_oemol("benzene")
molecule1 = generate_conformers(molecule1, max_confs=1)
molecule2 = generate_conformers(molecule2, max_confs=1)
# Allow ring breaking
new_to_old_atom_map = AtomMapper._get_mol_atom_map(
molecule1, molecule2, allow_ring_breaking=True)
if not len(new_to_old_atom_map) > 0:
filename = 'mapping-error.png'
#render_atom_mapping(filename, molecule1, molecule2, new_to_old_atom_map)
msg = 'Napthalene -> benzene transformation with allow_ring_breaking=True is not returning a valid mapping\n'
msg += 'Wrote atom mapping to %s for inspection; please check this.' % filename
msg += str(new_to_old_atom_map)
raise Exception(msg)
new_to_old_atom_map = AtomMapper._get_mol_atom_map(
molecule1, molecule2, allow_ring_breaking=False)
if new_to_old_atom_map is not None: # atom mapper should not retain _any_ atoms in default mode
filename = 'mapping-error.png'
#render_atom_mapping(filename, molecule1, molecule2, new_to_old_atom_map)
msg = 'Napthalene -> benzene transformation with allow_ring_breaking=False is erroneously allowing ring breaking\n'
msg += 'Wrote atom mapping to %s for inspection; please check this.' % filename
msg += str(new_to_old_atom_map)
raise Exception(msg)
def compare_energies(mol_name="naphthalene", ref_mol_name="benzene"):
"""
Make an atom map where the molecule at either lambda endpoint is identical, and check that the energies are also the same.
"""
from openmmtools import alchemy, states
from perses.rjmc.topology_proposal import SmallMoleculeSetProposalEngine, TopologyProposal
from perses.annihilation.relative import HybridTopologyFactory
import simtk.openmm as openmm
from perses.utils.openeye import createSystemFromIUPAC
from openmoltools.openeye import iupac_to_oemol, generate_conformers
mol = iupac_to_oemol(mol_name)
mol = generate_conformers(mol, max_confs=1)
m, system, positions, topology = createSystemFromIUPAC(mol_name)
refmol = iupac_to_oemol(ref_mol_name)
refmol = generate_conformers(refmol, max_confs=1)
#map one of the rings
atom_map = SmallMoleculeSetProposalEngine._get_mol_atom_map(mol, refmol)
#now use the mapped atoms to generate a new and old system with identical atoms mapped. This will result in the
#same molecule with the same positions for lambda=0 and 1, and ensures a contiguous atom map
effective_atom_map = {value: value for value in atom_map.values()}
#make a topology proposal with the appropriate data:
top_proposal = TopologyProposal(new_topology=topology,
new_system=system,
old_topology=topology,
old_system=system,
new_to_old_atom_map=effective_atom_map,
new_chemical_state_key="n1",
old_chemical_state_key='n2')
factory = HybridTopologyFactory(top_proposal, positions, positions)
alchemical_system = factory.hybrid_system
alchemical_positions = factory.hybrid_positions
platform = openmm.Platform.getPlatformByName("Reference")
_, _, alch_zero_state, alch_one_state = utils.generate_endpoint_thermodynamic_states(
alchemical_system, top_proposal)
rp_list = []
for state in [alch_zero_state, alch_one_state]:
integrator = openmm.VerletIntegrator(1)
context = state.create_context(integrator, platform)
samplerstate = states.SamplerState(
positions=alchemical_positions,
box_vectors=alchemical_system.getDefaultPeriodicBoxVectors())
samplerstate.apply_to_context(context)
rp = state.reduced_potential(context)
rp_list.append(rp)
del context, integrator
assert abs(rp_list[0] - rp_list[1]) < 1e-6
def createSystemFromIUPAC(iupac_name, title="MOL", **system_generator_kwargs):
"""
Create an openmm system out of an oemol
Parameters
----------
iupac_name : str
IUPAC name
Returns
-------
molecule : openeye.oechem.OEMol
OEMol molecule
system : openmm.System object
OpenMM system
positions : [n,3] np.array of floats
Positions
topology : openmm.app.Topology object
Topology
"""
# Create OEMol
# TODO write our own of this function so we can be
# sure of the oe flags that are being used
molecule = iupac_to_oemol(iupac_name, title=title)
molecule = generate_conformers(molecule, max_confs=1)
system_generator = system_generator_wrapper([molecule], **system_generator_kwargs)
# generate openmm system, positions and topology
system, positions, topology = OEMol_to_omm_ff(molecule, system_generator)
return (molecule, system, positions, topology)
json_data = {}
json_data["driver"] = "property"
json_data["kwargs"] = {"properties": ["WIBERG_LOWDIN_INDICES", "MAYER_INDICES"]}
json_data["method"] = "hf3c"
json_data["options"] = {"BASIS": "def2-svp"}
json_data["return_output"] = "True"
for mol in mollist[2:]:
name = mol.GetTitle()
print('\nGenerating input for {}'.format(name))
try:
os.mkdir(os.path.join(output_path, name))
except FileExistsError:
print("Overwriting existing directory")
conformers = openeye.generate_conformers(mol)
tagged_smiles = oechem.OEMolToSmiles(mol)
json_data["tagged_smiles"] = tagged_smiles
molecule, atom_map = utils.get_atom_map(tagged_smiles, conformers, is_mapped=True)
if not atom_map:
print("Can't get atom map. Skipping {}".format(name))
continue
print("{} has {} conformers".format(name, conformers.GetMaxConfIdx()))
# Generate mol2 file
openeye.molecule_to_mol2(conformers, tripos_mol2_filename='{}.mol2'.format(name), conformer=None)
# Generate xyz file
#xyz = utils.to_mapped_xyz(conformers, atom_map=atom_map, xyz_format=True, filename=name)
xyz = utils.to_mapped_xyz(conformers, atom_map=atom_map)
for i, coords in enumerate(xyz.split('*')):
json_data['molecule'] = coords
def __init__(self, molecules: List[str], output_filename: str, ncmc_switching_times: Dict[str, int], equilibrium_steps: Dict[str, int], timestep: unit.Quantity, initial_molecule: str=None, geometry_options: Dict=None):
self._molecules = [SmallMoleculeSetProposalEngine.canonicalize_smiles(molecule) for molecule in molecules]
environments = ['explicit', 'vacuum']
temperature = 298.15 * unit.kelvin
pressure = 1.0 * unit.atmospheres
constraints = app.HBonds
self._storage = NetCDFStorage(output_filename)
self._ncmc_switching_times = ncmc_switching_times
self._n_equilibrium_steps = equilibrium_steps
self._geometry_options = geometry_options
# Create a system generator for our desired forcefields.
from perses.rjmc.topology_proposal import SystemGenerator
system_generators = dict()
from pkg_resources import resource_filename
gaff_xml_filename = resource_filename('perses', 'data/gaff.xml')
barostat = openmm.MonteCarloBarostat(pressure, temperature)
system_generators['explicit'] = SystemGenerator([gaff_xml_filename, 'tip3p.xml'],
forcefield_kwargs={'nonbondedCutoff': 9.0 * unit.angstrom,
'implicitSolvent': None,
'constraints': constraints,
'ewaldErrorTolerance': 1e-5,
'hydrogenMass': 3.0*unit.amu}, periodic_forcefield_kwargs = {'nonbondedMethod': app.PME}
barostat=barostat)
system_generators['vacuum'] = SystemGenerator([gaff_xml_filename],
forcefield_kwargs={'implicitSolvent': None,
'constraints': constraints,
'hydrogenMass': 3.0*unit.amu}, nonperiodic_forcefield_kwargs = {'nonbondedMethod': app.NoCutoff})
#
# Create topologies and positions
#
topologies = dict()
positions = dict()
from openmoltools import forcefield_generators
forcefield = app.ForceField(gaff_xml_filename, 'tip3p.xml')
forcefield.registerTemplateGenerator(forcefield_generators.gaffTemplateGenerator)
# Create molecule in vacuum.
from perses.utils.openeye import extractPositionsFromOEMol
from openmoltools.openeye import smiles_to_oemol, generate_conformers
if initial_molecule:
smiles = initial_molecule
else:
smiles = np.random.choice(molecules)
molecule = smiles_to_oemol(smiles)
molecule = generate_conformers(molecule, max_confs=1)
topologies['vacuum'] = forcefield_generators.generateTopologyFromOEMol(molecule)
positions['vacuum'] = extractPositionsFromOEMol(molecule)
# Create molecule in solvent.
modeller = app.Modeller(topologies['vacuum'], positions['vacuum'])
modeller.addSolvent(forcefield, model='tip3p', padding=9.0 * unit.angstrom)
topologies['explicit'] = modeller.getTopology()
positions['explicit'] = modeller.getPositions()
# Set up the proposal engines.
proposal_metadata = {}
proposal_engines = dict()
for environment in environments:
proposal_engines[environment] = SmallMoleculeSetProposalEngine(self._molecules,
system_generators[environment])
# Generate systems
systems = dict()
for environment in environments:
systems[environment] = system_generators[environment].build_system(topologies[environment])
# Define thermodynamic state of interest.
thermodynamic_states = dict()
thermodynamic_states['explicit'] = states.ThermodynamicState(system=systems['explicit'],
temperature=temperature, pressure=pressure)
thermodynamic_states['vacuum'] = states.ThermodynamicState(system=systems['vacuum'], temperature=temperature)
# Create SAMS samplers
from perses.samplers.samplers import ExpandedEnsembleSampler, SAMSSampler
mcmc_samplers = dict()
exen_samplers = dict()
sams_samplers = dict()
for environment in environments:
storage = NetCDFStorageView(self._storage, envname=environment)
if self._geometry_options:
n_torsion_divisions = self._geometry_options['n_torsion_divsions'][environment]
use_sterics = self._geometry_options['use_sterics'][environment]
else:
n_torsion_divisions = 180
use_sterics = False
geometry_engine = geometry.FFAllAngleGeometryEngine(storage=storage, n_torsion_divisions=n_torsion_divisions, use_sterics=use_sterics)
move = mcmc.LangevinSplittingDynamicsMove(timestep=timestep, splitting="V R O R V",
n_restart_attempts=10)
chemical_state_key = proposal_engines[environment].compute_state_key(topologies[environment])
if environment == 'explicit':
sampler_state = states.SamplerState(positions=positions[environment],
box_vectors=systems[environment].getDefaultPeriodicBoxVectors())
else:
sampler_state = states.SamplerState(positions=positions[environment])
mcmc_samplers[environment] = mcmc.MCMCSampler(thermodynamic_states[environment], sampler_state, move)
exen_samplers[environment] = ExpandedEnsembleSampler(mcmc_samplers[environment], topologies[environment],
chemical_state_key, proposal_engines[environment],
geometry_engine,
options={'nsteps': self._ncmc_switching_times[environment]}, storage=storage, ncmc_write_interval=self._ncmc_switching_times[environment])
exen_samplers[environment].verbose = True
sams_samplers[environment] = SAMSSampler(exen_samplers[environment], storage=storage)
sams_samplers[environment].verbose = True
# Create test MultiTargetDesign sampler.
from perses.samplers.samplers import MultiTargetDesign
target_samplers = {sams_samplers['explicit']: 1.0, sams_samplers['vacuum']: -1.0}
designer = MultiTargetDesign(target_samplers, storage=self._storage)
# Store things.
self.molecules = molecules
self.environments = environments
self.topologies = topologies
self.positions = positions
self.system_generators = system_generators
self.proposal_engines = proposal_engines
self.thermodynamic_states = thermodynamic_states
self.mcmc_samplers = mcmc_samplers
self.exen_samplers = exen_samplers
self.sams_samplers = sams_samplers
self.designer = designer
def generate_torsions(inp_mol, output_path, interval, base_name=None, tar=True):
"""
This function takes a 3D molecule (pdf, mol2 or sd file) and generates structures for a torsion drive on all torsions
in the molecule. This function uses OpenEye
Parameters
----------
mol : OEMol
molecule to generate 1D torsion scans
output_path: str
path to output file directory
interval: int
angle (in degrees) of interval for torsion drive
base_name: str
base name for file. Default is None. If default, use title in OEMol for base name
tar: bool
If true, will compress output
"""
if not base_name:
base_name = inp_mol.GetTitle()
mid_tors = [[tor.a, tor.b, tor.c, tor.d ] for tor in oechem.OEGetTorsions(inp_mol)]
# This smarts should match terminal torsions such as -CH3, -NH2, -NH3+, -OH, and -SH
smarts = '[*]~[*]-[X2H1,X3H2,X4H3]-[#1]'
qmol=oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smarts):
warnings.warn('OEParseSmarts failed')
ss = oechem.OESubSearch(qmol)
mol = oechem.OEMol(inp_mol)
h_tors = []
oechem.OEPrepareSearch(mol, ss)
unique = True
for match in ss.Match(mol, unique):
tor = []
for ma in match.GetAtoms():
tor.append(ma.target)
h_tors.append(tor)
# Combine middle and terminal torsions
all_tors = mid_tors + h_tors
# Sort all_tors so that it's grouped by central bond
central_bonds = np.zeros((len(all_tors), 3), dtype=int)
for i, tor in enumerate(all_tors):
central_bonds[i][0] = i
central_bonds[i][1] = tor[1].GetIdx()
central_bonds[i][2] = tor[2].GetIdx()
grouped = central_bonds[central_bonds[:, 2].argsort()]
sorted_tors = [all_tors[i] for i in grouped[:, 0]]
# Keep only one torsion per rotatable bond
tors = []
best_tor = [sorted_tors[0][0], sorted_tors[0][0], sorted_tors[0][0], sorted_tors[0][0]]
first_pass = True
for tor in sorted_tors:
logger().info("Idxs: {} {} {} {}".format(tor[0].GetIdx(), tor[1].GetIdx(), tor[2].GetIdx(), tor[3].GetIdx()))
logger().info("Atom Numbers: {} {} {} {}".format(tor[0].GetAtomicNum(), tor[1].GetAtomicNum(), tor[2].GetAtomicNum(), tor[3].GetAtomicNum()))
if tor[1].GetIdx() != best_tor[1].GetIdx() or tor[2].GetIdx() != best_tor[2].GetIdx():
new_tor = True
if not first_pass:
logger().info("Adding to list: {} {} {} {}".format(best_tor[0].GetIdx(), best_tor[1].GetIdx(), best_tor[2].GetIdx(), best_tor[3].GetIdx()))
tors.append(best_tor)
first_pass = False
best_tor = tor
best_tor_order = tor[0].GetAtomicNum() + tor[3].GetAtomicNum()
logger().info("new_tor with central bond across atoms: {} {}".format(tor[1].GetIdx(), tor[2].GetIdx()))
else:
logger().info("Not a new_tor but now with end atoms: {} {}".format(tor[0].GetIdx(), tor[3].GetIdx()))
tor_order = tor[0].GetAtomicNum() + tor[3].GetAtomicNum()
if tor_order > best_tor_order:
best_tor = tor
best_tor_order = tor_order
logger().info("Adding to list: {} {} {} {}".format(best_tor[0].GetIdx(), best_tor[1].GetIdx(), best_tor[2].GetIdx(), best_tor[3].GetIdx()))
tors.append(best_tor)
logger().info("List of torsion to drive:")
for tor in tors:
logger().info("Idx: {} {} {} {}".format(tor[0].GetIdx(), tor[1].GetIdx(), tor[2].GetIdx(), tor[3].GetIdx()))
logger().info("Atom numbers: {} {} {} {}".format(tor[0].GetAtomicNum(), tor[1].GetAtomicNum(), tor[2].GetAtomicNum(), tor[3].GetAtomicNum()))
conf = mol.GetConfs().next()
coords = oechem.OEFloatArray(conf.GetMaxAtomIdx() * 3)
conf.GetCoords(coords)
# Check if coordinates are not zero
values = np.asarray([coords.__getitem__(i) == 0 for i in range(coords.__len__())])
if values.all():
# Generate new coordinates.
mol2 = generate_conformers(mol, max_confs=1)
conf = mol2.GetConfs().next()
coords = oechem.OEFloatArray(conf.GetMaxAtomIdx() * 3)
conf.GetCoords(coords)
mol2.DeleteConfs()
mol.DeleteConfs()
for tor in tors:
tor_name = str((tor[0].GetIdx())+1) + '_' + str((tor[1].GetIdx())+1) + '_' + str((tor[2].GetIdx())+1) + '_' + str((tor[3].GetIdx())+1)
folder = os.path.join(output_path, tor_name)
try:
os.makedirs(folder)
except FileExistsError:
logger().info("Overwriting existing directory {}".format(tor_name))
for angle in range(0, 360, interval):
angle_folder = os.path.join(folder, str(angle))
try:
os.mkdir(angle_folder)
except FileExistsError:
logger().info("Overwriting existing directory {}".format(tor_name))
newconf = mol.NewConf(coords)
oechem.OESetTorsion(newconf, tor[0], tor[1], tor[2], tor[3], radians(angle))
pdb = oechem.oemolostream('{}/{}_{}_{}.pdb'.format(angle_folder, base_name, tor_name, angle))
oechem.OEWritePDBFile(pdb, newconf)
if tar:
# tar archive output
out = tarfile.open('{}.tar.gz'.format(output_path), mode='w:gz')
os.chdir(output_path)
os.chdir('../')
out.add('{}'.format(base_name))
out.close()
def generate_torsions(inp_mol,
output_path,
interval,
base_name=None,
tar=True):
"""
This function takes a 3D molecule (pdf, mol2 or sd file) and generates structures for a torsion drive on all torsions
in the molecule. This function uses OpenEye
Parameters
----------
mol : OEMol
molecule to generate 1D torsion scans
output_path: str
path to output file directory
interval: int
angle (in degrees) of interval for torsion drive
base_name: str
base name for file. Default is None. If default, use title in OEMol for base name
tar: bool
If true, will compress output
"""
if not base_name:
base_name = inp_mol.GetTitle()
mid_tors = [[tor.a, tor.b, tor.c, tor.d]
for tor in oechem.OEGetTorsions(inp_mol)]
# This smarts should match terminal torsions such as -CH3, -NH2, -NH3+, -OH, and -SH
smarts = '[*]~[*]-[X2H1,X3H2,X4H3]-[#1]'
qmol = oechem.OEQMol()
if not oechem.OEParseSmarts(qmol, smarts):
warnings.warn('OEParseSmarts failed')
ss = oechem.OESubSearch(qmol)
mol = oechem.OEMol(inp_mol)
h_tors = []
oechem.OEPrepareSearch(mol, ss)
unique = True
for match in ss.Match(mol, unique):
tor = []
for ma in match.GetAtoms():
tor.append(ma.target)
h_tors.append(tor)
# Combine middle and terminal torsions
all_tors = mid_tors + h_tors
# Sort all_tors so that it's grouped by central bond
central_bonds = np.zeros((len(all_tors), 3), dtype=int)
for i, tor in enumerate(all_tors):
central_bonds[i][0] = i
central_bonds[i][1] = tor[1].GetIdx()
central_bonds[i][2] = tor[2].GetIdx()
grouped = central_bonds[central_bonds[:, 2].argsort()]
sorted_tors = [all_tors[i] for i in grouped[:, 0]]
# Keep only one torsion per rotatable bond
tors = []
best_tor = [
sorted_tors[0][0], sorted_tors[0][0], sorted_tors[0][0],
sorted_tors[0][0]
]
first_pass = True
for tor in sorted_tors:
logger().info("Idxs: {} {} {} {}".format(tor[0].GetIdx(),
tor[1].GetIdx(),
tor[2].GetIdx(),
tor[3].GetIdx()))
logger().info("Atom Numbers: {} {} {} {}".format(
tor[0].GetAtomicNum(), tor[1].GetAtomicNum(),
tor[2].GetAtomicNum(), tor[3].GetAtomicNum()))
if tor[1].GetIdx() != best_tor[1].GetIdx() or tor[2].GetIdx(
) != best_tor[2].GetIdx():
new_tor = True
if not first_pass:
logger().info("Adding to list: {} {} {} {}".format(
best_tor[0].GetIdx(), best_tor[1].GetIdx(),
best_tor[2].GetIdx(), best_tor[3].GetIdx()))
tors.append(best_tor)
first_pass = False
best_tor = tor
best_tor_order = tor[0].GetAtomicNum() + tor[3].GetAtomicNum()
logger().info(
"new_tor with central bond across atoms: {} {}".format(
tor[1].GetIdx(), tor[2].GetIdx()))
else:
logger().info("Not a new_tor but now with end atoms: {} {}".format(
tor[0].GetIdx(), tor[3].GetIdx()))
tor_order = tor[0].GetAtomicNum() + tor[3].GetAtomicNum()
if tor_order > best_tor_order:
best_tor = tor
best_tor_order = tor_order
logger().info("Adding to list: {} {} {} {}".format(best_tor[0].GetIdx(),
best_tor[1].GetIdx(),
best_tor[2].GetIdx(),
best_tor[3].GetIdx()))
tors.append(best_tor)
logger().info("List of torsion to drive:")
for tor in tors:
logger().info("Idx: {} {} {} {}".format(tor[0].GetIdx(),
tor[1].GetIdx(),
tor[2].GetIdx(),
tor[3].GetIdx()))
logger().info("Atom numbers: {} {} {} {}".format(
tor[0].GetAtomicNum(), tor[1].GetAtomicNum(),
tor[2].GetAtomicNum(), tor[3].GetAtomicNum()))
conf = mol.GetConfs().next()
coords = oechem.OEFloatArray(conf.GetMaxAtomIdx() * 3)
conf.GetCoords(coords)
# Check if coordinates are not zero
values = np.asarray(
[coords.__getitem__(i) == 0 for i in range(coords.__len__())])
if values.all():
# Generate new coordinates.
mol2 = generate_conformers(mol, max_confs=1)
conf = mol2.GetConfs().next()
coords = oechem.OEFloatArray(conf.GetMaxAtomIdx() * 3)
conf.GetCoords(coords)
mol2.DeleteConfs()
mol.DeleteConfs()
for tor in tors:
tor_name = str((tor[0].GetIdx()) + 1) + '_' + str(
(tor[1].GetIdx()) + 1) + '_' + str(
(tor[2].GetIdx()) + 1) + '_' + str((tor[3].GetIdx()) + 1)
folder = os.path.join(output_path, tor_name)
try:
os.makedirs(folder)
except FileExistsError:
logger().info("Overwriting existing directory {}".format(tor_name))
for angle in range(0, 360, interval):
angle_folder = os.path.join(folder, str(angle))
try:
os.mkdir(angle_folder)
except FileExistsError:
logger().info(
"Overwriting existing directory {}".format(tor_name))
newconf = mol.NewConf(coords)
oechem.OESetTorsion(newconf, tor[0], tor[1], tor[2], tor[3],
radians(angle))
pdb = oechem.oemolostream('{}/{}_{}_{}.pdb'.format(
angle_folder, base_name, tor_name, angle))
oechem.OEWritePDBFile(pdb, newconf)
if tar:
# tar archive output
out = tarfile.open('{}.tar.gz'.format(output_path), mode='w:gz')
os.chdir(output_path)
os.chdir('../')
out.add('{}'.format(base_name))
out.close()