def get_iupac(molecule):
"""
Generate IUPAC name
Parameters
----------
molecule :
`oechem.OEMol`
Returns
-------
str:
iupac name
Notes
-----
Will only be generated if has openeye license
"""
if not has_openeye:
raise ImportError(
"OpenEye is not installed. You can use the canonicalization='rdkit' to use the RDKit backend"
"The Conda recipe for cmiles installs rdkit")
from openeye import oeiupac
if not oeiupac.OEIUPACIsLicensed():
raise ImportError("Must have OEIUPAC license!")
return oeiupac.OECreateIUPACName(molecule)
def _create_implicit_solvent_openmm(self, mol):
"""
Take a list of oemols, and generate openmm systems
and positions for each.
Parameters
----------
mol : oemol
oemol to be turned into system, positions
Returns
-------
system : simtk.openmm.System
openmm system corresponding to molecule
positions : np.array, Quantity nm
array of atomic positions
"""
molecule_name = oeiupac.OECreateIUPACName(mol)
openmoltools.openeye.enter_temp_directory()
_, tripos_mol2_filename = openmoltools.openeye.molecule_to_mol2(
mol,
tripos_mol2_filename=molecule_name + '.tripos.mol2',
conformer=0,
residue_name='MOL')
gaff_mol2, frcmod = openmoltools.amber.run_antechamber(
molecule_name, tripos_mol2_filename)
prmtop_file, inpcrd_file = openmoltools.amber.run_tleap(
molecule_name, gaff_mol2, frcmod)
prmtop = app.AmberPrmtopFile(prmtop_file)
crd = app.AmberInpcrdFile(inpcrd_file)
system = prmtop.createSystem(implicitSolvent=self.implicit_solvent,
constraints=self.constraints,
removeCMMotion=False)
positions = crd.getPositions(asNumpy=True)
return system, positions
def generate_molecule_from_smiles(smiles, name=None):
"""
Parameters
----------
smiles : str
The canonical isomeric SMILES string.
name : str, optional, default=None
If specified, the molecule title will be set to this; if not, the IUPAC name will be assigned.
"""
# Generate a molecule from canonical isomeric SMILES.
molecule = oechem.OEMol()
if not oechem.OEParseSmiles(molecule, smiles):
raise ValueError("The supplied SMILES '%s' could not be parsed." %
smiles)
# Assign aromaticity.
oechem.OEAssignAromaticFlags(molecule, oechem.OEAroModelOpenEye)
# Add hydrogens.
oechem.OEAddExplicitHydrogens(molecule)
# Set title.
if name is None:
# Set title to IUPAC name.
name = oeiupac.OECreateIUPACName(molecule)
molecule.SetTitle(name)
# Check for any missing atom names, if found reassign all of them.
if any([atom.GetName() == '' for atom in molecule.GetAtoms()]):
oechem.OETriposAtomNames(molecule)
return molecule
def normalize_molecule(molecule):
"""Normalize a copy of the molecule by checking aromaticity, adding explicit hydrogens, and renaming by IUPAC name.
Parameters
----------
molecule : OEMol
the molecule to be normalized.
Returns
-------
molcopy : OEMol
A (copied) version of the normalized molecule
"""
molcopy = oechem.OEMol(molecule)
# Assign aromaticity.
oechem.OEAssignAromaticFlags(molcopy, oechem.OEAroModelOpenEye)
# Add hydrogens.
oechem.OEAddExplicitHydrogens(molcopy)
# Set title to IUPAC name.
name = oeiupac.OECreateIUPACName(molcopy)
molcopy.SetTitle(name)
# Check for any missing atom names, if found reassign all of them.
if any([atom.GetName() == '' for atom in molcopy.GetAtoms()]):
oechem.OETriposAtomNames(molcopy)
return molcopy
def __makeChemCompIdentifierCategory(self, ccId, oeMol):
"""
loop_
_pdbx_chem_comp_identifier.comp_id
_pdbx_chem_comp_identifier.type
_pdbx_chem_comp_identifier.program
_pdbx_chem_comp_identifier.program_version
_pdbx_chem_comp_identifier.identifier
ATP "SYSTEMATIC NAME" ACDLabs 10.04
;adenosine 5'-(tetrahydrogen triphosphate)
;
ATP "SYSTEMATIC NAME" "OpenEye OEToolkits" 1.5.0 "[[(2R,3S,4R,5R)-5-(6-aminopurin-9-yl)-..."
#
"""
rowL = []
#
aRow = {}
aRow["comp_id"] = ccId
aRow["type"] = "SYSTEMATIC NAME"
aRow["program"] = "OpenEye OEToolkits"
aRow["program_version"] = self.__oeVersion
style = oeiupac.OEGetIUPACNamStyle("systematic")
name = oeiupac.OEToUTF8(oeiupac.OECreateIUPACName(oeMol, style))
aRow["identifier"] = name
rowL.append(aRow)
aRow = {}
aRow["comp_id"] = ccId
aRow["type"] = "COMMON"
aRow["program"] = "OpenEye OEToolkits"
aRow["program_version"] = self.__oeVersion
style = oeiupac.OEGetIUPACNamStyle("traditional")
name = oeiupac.OEToUTF8(oeiupac.OECreateIUPACName(oeMol, style))
aRow["identifier"] = name
rowL.append(aRow)
#
aRow = {}
aRow["comp_id"] = ccId
aRow["type"] = "SYNONYM"
aRow["program"] = "OpenEye OEToolkits"
aRow["program_version"] = self.__oeVersion
style = oeiupac.OEGetIUPACNamStyle("acdname")
name = oeiupac.OEToUTF8(oeiupac.OECreateIUPACName(oeMol, style))
aRow["identifier"] = name
rowL.append(aRow)
#
return rowL
def __makeChemCompCategory(self, ccId, oeMol, site="RCSB", missingModelXyz=False, skipAnnotations=False):
#
lt = time.strftime("%Y-%m-%d", time.localtime())
formula = oechem.OEMolecularFormula(oeMol)
charge = self.__getFormalCharge(oeMol)
fW = oechem.OECalculateMolecularWeight(oeMol)
#
if skipAnnotations:
name = ccId
else:
style = oeiupac.OEGetIUPACNamStyle("systematic")
name = oeiupac.OEToUTF8(oeiupac.OECreateIUPACName(oeMol, style))
#
ccRow = {}
ccRow["id"] = ccId
if name is not None:
ccRow["name"] = name
else:
ccRow["name"] = "?"
ccRow["type"] = "NON-POLYMER"
ccRow["pdbx_type"] = "?"
if formula is not None:
ccRow["formula"] = formula
else:
ccRow["formula"] = "?"
ccRow["mon_nstd_parent_comp_id"] = "?"
# ccRow["pdbx_synonyms"] = "?"
if charge is not None:
ccRow["pdbx_formal_charge"] = charge
else:
ccRow["pdbx_formal_charge"] = "?"
ccRow["pdbx_ambiguous_flag"] = "N"
ccRow["pdbx_initial_date"] = lt
ccRow["pdbx_modified_date"] = lt
ccRow["pdbx_release_status"] = "HOLD"
ccRow["pdbx_replaced_by"] = "?"
ccRow["pdbx_replaces"] = "?"
if fW is not None:
ccRow["formula_weight"] = "%0.3f" % fW
else:
ccRow["formula_weight"] = "?"
ccRow["one_letter_code"] = "?"
tlc = ccId.split("_")[0]
ccRow["three_letter_code"] = tlc
ccRow["pdbx_model_coordinates_details"] = "?"
ccRow["pdbx_ideal_coordinates_details"] = "?"
if missingModelXyz:
ccRow["pdbx_model_coordinates_missing_flag"] = "Y"
else:
ccRow["pdbx_model_coordinates_missing_flag"] = "N"
ccRow["pdbx_model_coordinates_db_code"] = "?"
ccRow["pdbx_processing_site"] = site
ccRow["pdbx_subcomponent_list"] = "?"
return ccRow
def Mol2Nam(itf):
ifs = oechem.oemolistream()
if not ifs.open(itf.GetString("-in")):
oechem.OEThrow.Fatal("Unable to open '%s' for reading" % itf.GetString("-in"))
ofs = oechem.oemolostream()
outname = None
if itf.HasString("-out"):
outname = itf.GetString("-out")
if not ofs.open(outname):
oechem.OEThrow.Fatal("Unable to open '%s' for reading" % outname)
language = oeiupac.OEGetIUPACLanguage(itf.GetString("-language"))
charset = oeiupac.OEGetIUPACCharSet(itf.GetString("-encoding"))
style = oeiupac.OEGetIUPACNamStyle(itf.GetString("-style"))
for mol in ifs.GetOEGraphMols():
name = oeiupac.OECreateIUPACName(mol, style)
if language > 0:
name = oeiupac.OEToLanguage(name, language)
if itf.GetBool("-capitalize"):
name = oeiupac.OECapitalizeName(name)
if charset == oeiupac.OECharSet_ASCII:
name = oeiupac.OEToAscii(name)
elif charset == oeiupac.OECharSet_UTF8:
name = oeiupac.OEToUTF8(name)
elif charset == oeiupac.OECharSet_HTML:
name = oeiupac.OEToHTML(name)
elif charset == oeiupac.OECharSet_SJIS:
name = oeiupac.OEToSJIS(name)
elif charset == oeiupac.OECharSet_EUCJP:
name = oeiupac.OEToEUCJP(name)
if outname:
if itf.HasString("-delim"):
title = mol.GetTitle()
name = title + itf.GetString("-delim") + name
if itf.HasString("-tag"):
oechem.OESetSDData(mol, itf.GetString("-tag"), name)
mol.SetTitle(name)
oechem.OEWriteMolecule(ofs, mol)
else:
print(name)
def _state_transition_to_iupac(self, state_transition):
"""
Convenience function to convert SMILES to IUPAC names
Parameters
----------
state_transition : (str, str)
Pair of smiles strings for the state transition
Returns
-------
state_transition_iupac : [str, str]
The pair of molecules in IUPAC names
"""
state_transition_iupac = []
for state in state_transition:
mol = oechem.OEMol()
oechem.OESmilesToMol(mol, state)
iupac = oeiupac.OECreateIUPACName(mol)
state_transition_iupac.append(iupac)
return state_transition_iupac
template_script.format(yaml_filename)
if __name__ == "__main__":
template_script_file_eq = "submit-eq.sh"
template_script_file_neq = "submit-neq.sh"
substituted_benzene_smilefile = "filtered_database.smi"
substituted_benzenes_iupac = []
istream = oechem.oemolistream(substituted_benzene_smilefile)
for mol in istream.GetOEMols():
mol_copy = oechem.OEMol(mol)
substituted_benzenes_iupac.append(oeiupac.OECreateIUPACName(mol_copy))
with open("rj_neq_template.yaml", "r") as yamlfile:
template_yamldict = yaml.load(yamlfile)
for pair in itertools.permutations(substituted_benzenes_iupac, 2):
if pair[0] == pair[1]:
continue
new_yaml_dict = create_yaml_file(pair[0], pair[1],
copy.deepcopy(template_yamldict))
new_yaml_filename = "{}_{}_rjneq.yaml".format(pair[0], pair[1])
with open(new_yaml_filename, 'w') as yaml_outfile:
yaml.dump(new_yaml_dict, yaml_outfile)
def generate_fragments(molecule, generate_visualization=False, strict_stereo=False, combinatorial=True, MAX_ROTORS=2,
remove_map=True, json_filename=None):
"""
This function generates fragments from molecules. The output is a dictionary that maps SMILES of molecules to SMILES
for fragments. The default SMILES are generated with openeye.oechem.OEMolToSmiles. These SMILES strings are canonical
isomeric SMILES.
The dictionary also includes a provenance field which defines how the fragments were generated.
Parameters
----------
molecule: OEMol to fragment
generate_visualization: bool
If true, visualization of the fragments will be written to pdf files. The pdf will be writtten in the directory
where this function is run from.
combinatorial: bool
If true, find all connected fragments from fragments and add all new fragments that have less than MAX_ROTORS
MAX_ROTORS: int
rotor threshold for combinatorial
strict_stereo: bool
Note: This applies to the molecule being fragmented. Not the fragments.
If True, omega will generate conformation with stereochemistry defined in the SMILES string for charging.
remove_map: bool
If True, the index tags will be removed. This will remove duplicate fragments. Defualt True
json_filename: str
filenmae for JSON. If provided, will save the returned dictionary to a JSON file. Default is None
Returns
-------
fragments: dict
mapping of SMILES from the parent molecule to the SMILES of the fragments
"""
fragments = dict()
try:
molecules = list(molecule)
except TypeError:
molecules = [molecule]
for molecule in molecules:
# normalize molecule
molecule = normalize_molecule(molecule, molecule.GetTitle())
if remove_map:
# Remove tags from smiles. This is done to make it easier to find duplicate fragments
for a in molecule.GetAtoms():
a.SetMapIdx(0)
frags = _generate_fragments(molecule, strict_stereo=strict_stereo)
if not frags:
logger().warning('Skipping {}, SMILES: {}'.format(molecule.GetTitle(), oechem.OECreateSmiString(molecule)))
continue
charged = frags[0]
frags = frags[-1]
frag_list = list(frags.values())
if combinatorial:
smiles = smiles_with_combined(frag_list, charged, MAX_ROTORS)
else:
smiles = frag_to_smiles(frag_list, charged)
parent_smiles = mol_to_smiles(molecule, isomeric=True, explicit_hydrogen=False, mapped=False)
if smiles:
fragments[parent_smiles] = list(smiles.keys())
else:
# Add molecule where no fragments were found for terminal torsions and / or rings and non rotatable bonds
fragments[parent_smiles] = [mol_to_smiles(molecule, isomeric=True, explicit_hydrogen=True, mapped=False)]
if generate_visualization:
IUPAC = oeiupac.OECreateIUPACName(molecule)
name = molecule.GetTitle()
if IUPAC == name:
name = make_python_identifier(oechem.OEMolToSmiles(molecule))[0]
oname = '{}.pdf'.format(name)
ToPdf(charged, oname, frags)
del charged, frags
if json_filename:
f = open(json_filename, 'w')
j = json.dump(fragments, f, indent=2, sort_keys=True)
f.close()
return fragments