def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
# @ <SNIPPET-ALTLOCFACT-FLAVOR>
ims = oechem.oemolistream()
ims.SetFlavor(oechem.OEFormat_PDB, oechem.OEIFlavor_PDB_ALTLOC)
# @ </SNIPPET-ALTLOCFACT-FLAVOR>
inputFile = itf.GetString("-in")
if not ims.open(inputFile):
oechem.OEThrow.Fatal("Unable to open %s for reading." % inputFile)
if not oechem.OEIs3DFormat(ims.GetFormat()):
oechem.OEThrow.Fatal("%s is not in a 3D format." % inputFile)
mol = oechem.OEGraphMol()
if not oechem.OEReadMolecule(ims, mol):
oechem.OEThrow.Fatal("Unable to read %s." % inputFile)
if not oechem.OEHasResidues(mol):
oechem.OEPerceiveResidues(mol, oechem.OEPreserveResInfo_All)
# @ <SNIPPET-ALTLOCFACT-PRIMARY>
alf = oechem.OEAltLocationFactory(mol)
if alf.GetGroupCount() != 0:
alf.MakePrimaryAltMol(mol)
# @ </SNIPPET-ALTLOCFACT-PRIMARY>
# @ <SNIPPET-PLACE-HYDROGENS-BASIC>
if oechem.OEPlaceHydrogens(mol):
# ...
# @ </SNIPPET-PLACE-HYDROGENS-BASIC>
print("success")
# @ <SNIPPET-PLACE-HYDROGENS-OPTIONS>
opt = oechem.OEPlaceHydrogensOptions()
opt.SetStandardizeBondLen(False)
# @ </SNIPPET-PLACE-HYDROGENS-OPTIONS>
# @ <SNIPPET-PLACE-HYDROGENS-DETAILS>
# given molecule mol and OEPlaceHydrogensOptions opt...
details = oechem.OEPlaceHydrogensDetails()
if oechem.OEPlaceHydrogens(mol, details, opt):
print(details.Describe())
# @ </SNIPPET-PLACE-HYDROGENS-DETAILS>
ims.close()
def addh(self, altProcess='occupancy', processName='fullsearch', \
ihopt=T, standardize=T, badclash=0.4, flipbias=1.0, maxStates=20):#7):
imol = self.mol.CreateCopy()
wp = oechem.OEPlaceHydrogensWaterProcessing_Ignore
if processName == 'fullsearch':
wp = oechem.OEPlaceHydrogensWaterProcessing_FullSearch
elif processName == 'focused':
wp = oechem.OEPlaceHydrogensWaterProcessing_Focused
keepAlts = (altProcess != "a")
highestOcc = (altProcess == "occupancy")
compareAlts = (altProcess == "compare")
print('#1')
if highestOcc or compareAlts:
alf = oechem.OEAltLocationFactory(imol)
if alf.GetGroupCount() != 0:
if highestOcc:
oechem.OEThrow.Verbose(
"Dropping alternate locations from input.")
alf.MakePrimaryAltMol(imol)
elif compareAlts:
oechem.OEThrow.Verbose("Fixing alternate location issues.")
imol = alf.GetSourceMol()
omol = imol
print('#2')
oechem.OEThrow.Verbose("Adding hydrogens to complex.")
hopt = oechem.OEPlaceHydrogensOptions()
if ihopt:
hopt.SetAltsMustBeCompatible(compareAlts)
hopt.SetStandardizeBondLen(standardize)
hopt.SetWaterProcessing(wp)
hopt.SetBadClashOverlapDistance(badclash)
hopt.SetFlipBiasScale(flipbias)
hopt.SetMaxSubstateCutoff(maxStates)
if self.verbose:
print('#3')
details = oechem.OEPlaceHydrogensDetails()
if not oechem.OEPlaceHydrogens(omol, details, hopt):
oechem.OEThrow.Fatal(
"Unable to place hydrogens and get details on %s." %
self.inmol.GetTitle())
oechem.OEThrow.Verbose(details.Describe())
else:
if not oechem.OEPlaceHydrogens(omol, hopt):
oechem.OEThrow.Fatal("Unable to place hydrogens on %s." %
self.inmol.GetTitle())
self.mol = omol
def _prepare_ligand_template(
self, ligand_template: pd.Series, kinase_domain: oechem.OEGraphMol
) -> oechem.OEGraphMol:
"""
Prepare a PDB structure containing the ligand template of interest.
Parameters
----------
ligand_template: pd.Series
A data series containing entries 'pdb', 'chain', 'ligand' and 'alt'.
kinase_domain: oechem.OEGraphMol
An OpenEye molecule holding the kinase domain the ligand template structure should be superposed to.
Returns
-------
: oechem.OEGraphMol
An OpenEye molecule holding the prepared ligand structure.
"""
from openeye import oechem
from ..modeling.OEModeling import (
read_molecules,
select_chain,
select_altloc,
remove_non_protein,
superpose_proteins,
)
from ..utils import FileDownloader
logging.debug("Interpreting structure ...")
ligand_template_structure = PDBProtein(ligand_template.pdb)
if not ligand_template_structure.path.is_file():
logging.debug(f"Downloading PDB entry {ligand_template.pdb} ...")
FileDownloader.rcsb_structure_pdb(ligand_template.pdb)
logging.debug("Reading structure ...")
ligand_template_structure = read_molecules(ligand_template_structure.path)[0]
logging.debug("Selecting chain ...")
ligand_template_structure = select_chain(ligand_template_structure, ligand_template.chain)
logging.debug("Selecting alternate location ...")
ligand_template_structure = select_altloc(ligand_template_structure, ligand_template.alt)
logging.debug("Removing everything but protein, water and ligand of interest ...")
ligand_template_structure = remove_non_protein(
ligand_template_structure, exceptions=[ligand_template.ligand], remove_water=False
)
logging.debug("Superposing structure on kinase domain ...")
ligand_template_structure = superpose_proteins(kinase_domain, ligand_template_structure)
logging.debug("Adding hydrogens ...")
oechem.OEPlaceHydrogens(ligand_template_structure)
split_options = oechem.OESplitMolComplexOptions()
logging.debug("Extracting ligand ...")
ligand_template_structure = list(
oechem.OEGetMolComplexComponents(
ligand_template_structure, split_options, split_options.GetLigandFilter()
)
)[0]
return ligand_template_structure
def read_graph_mol(fname):
ifs = oechem.oemolistream(fname)
mol = oechem.OEGraphMol()
with oechem.oemolistream(fname) as ifs:
oechem.OEReadMolecule(ifs, mol)
oechem.OEPlaceHydrogens(mol)
return mol
def _assemble_complex(protein: oechem.OEGraphMol,
solvent: oechem.OEGraphMol,
ligand: oechem.OEGraphMol) -> oechem.OEGraphMol:
"""
Assemble components of a solvated protein-ligand complex into a single
OpenEye molecule. Water molecules will be checked for clashes with the ligand.
Parameters
----------
protein: oechem.OEGraphMol
An OpenEye molecule holding the protein of interest.
solvent: oechem.OEGraphMol
An OpenEye molecule holding the solvent of interest.
ligand: oechem.OEGraphMol
An OpenEye molecule holding the ligand of interest.
Returns
-------
protein_ligand_complex: oechem.OEGraphMol
An OpenEye molecule holding protein, ligand and solvent.
"""
from openeye import oechem
from ..modeling.OEModeling import (
clashing_atoms,
update_residue_identifiers,
split_molecule_components,
)
protein_ligand_complex = oechem.OEGraphMol()
logging.debug("Adding protein ...")
oechem.OEAddMols(protein_ligand_complex, protein)
logging.debug("Adding ligand ...")
oechem.OEAddMols(protein_ligand_complex, ligand)
logging.debug("Adding water molecules ...")
# converts solvent molecule into a list of water molecules
water_molecules = split_molecule_components(solvent)
for water_molecule in water_molecules: # TODO: slow, move to cKDtrees
if not clashing_atoms(ligand, water_molecule):
oechem.OEAddMols(protein_ligand_complex, water_molecule)
logging.debug("Updating hydrogen positions ...")
oechem.OEPlaceHydrogens(protein_ligand_complex)
logging.debug("Updating residue identifiers ...")
protein_ligand_complex = update_residue_identifiers(
protein_ligand_complex)
return protein_ligand_complex
def main(argv=[__name__]):
itf = oechem.OEInterface(InterfaceData)
if not oechem.OEParseCommandLine(itf, argv):
oechem.OEThrow.Fatal("Unable to interpret command line!")
ims = oechem.oemolistream()
ims.SetFlavor(oechem.OEFormat_PDB, oechem.OEIFlavor_PDB_ALTLOC)
inputFile = itf.GetString("-in")
if not ims.open(inputFile):
oechem.OEThrow.Fatal("Unable to open %s for reading." % inputFile)
if not oechem.OEIs3DFormat(ims.GetFormat()):
oechem.OEThrow.Fatal("%s is not in a 3D format." % inputFile)
mol = oechem.OEGraphMol()
if not oechem.OEReadMolecule(ims, mol):
oechem.OEThrow.Fatal("Unable to read %s." % inputFile)
if not oechem.OEHasResidues(mol):
oechem.OEPerceiveResidues(mol, oechem.OEPreserveResInfo_All)
alf = oechem.OEAltLocationFactory(mol)
if alf.GetGroupCount() != 0:
alf.MakePrimaryAltMol(mol)
# in our example, we will select the first histidine
selectedResidue = oechem.OEResidue()
for atom in mol.GetAtoms():
res = oechem.OEAtomGetResidue(atom)
if oechem.OEGetResidueIndex(res) == oechem.OEResidueIndex_HIS:
selectedResidue = res
break
# @ <SNIPPET-PLACE-HYDROGENS-PRED>
# given predicate IsSameResidue, residue selectedResidue and molecule mol...
opt = oechem.OEPlaceHydrogensOptions()
opt.SetNoFlipPredicate(IsSameResidue(selectedResidue))
if oechem.OEPlaceHydrogens(mol, opt):
# selectedResidue will not be flipped...
# @ </SNIPPET-PLACE-HYDROGENS-PRED>
print("success")
ims.close()
def mutate_structure(target_structure: oechem.OEGraphMol,
template_sequence: str) -> oechem.OEGraphMol:
"""
Mutate a protein structure according to an amino acid sequence.
Parameters
----------
target_structure: oechem.OEGraphMol
An OpenEye molecule holding a protein structure to mutate.
template_sequence: str
A template one letter amino acid sequence, which defines the sequence the target structure should be mutated
to. Protein residues not matching a template sequence will be either mutated or deleted.
Returns
-------
mutated_structure: oechem.OEGraphMol
An OpenEye molecule holding the mutated protein structure.
"""
from Bio import pairwise2
# the hierarchy view is more stable if reinitialized after each change
# https://docs.eyesopen.com/toolkits/python/oechemtk/biopolymers.html#a-hierarchy-view
finished = False
while not finished:
altered = False
# align template and target sequences
target_sequence = get_sequence(target_structure)
template_sequence_aligned, target_sequence_aligned = pairwise2.align.globalxs(
template_sequence, target_sequence, -10, 0)[0][:2]
logging.debug(f"Template sequence:\n{template_sequence}")
logging.debug(f"Target sequence:\n{target_sequence}")
hierview = oechem.OEHierView(target_structure)
structure_residues = hierview.GetResidues()
# adjust target structure to match template sequence
for template_sequence_residue, target_sequence_residue in zip(
template_sequence_aligned, target_sequence_aligned):
if template_sequence_residue == "-":
# delete any non protein residue from target structure
structure_residue = structure_residues.next()
if target_sequence_residue != "X":
# delete
for atom in structure_residue.GetAtoms():
target_structure.DeleteAtom(atom)
# break loop and reinitialize
altered = True
break
else:
# compare amino acids
if target_sequence_residue != "-":
structure_residue = structure_residues.next()
if target_sequence_residue not in [
"X", template_sequence_residue
]:
# mutate
structure_residue = structure_residue.GetOEResidue()
three_letter_code = oechem.OEGetResidueName(
oechem.OEGetResidueIndexFromCode(
template_sequence_residue))
oespruce.OEMutateResidue(target_structure,
structure_residue,
three_letter_code)
# break loop and reinitialize
altered = True
break
# leave while loop if no changes were introduced
if not altered:
finished = True
# OEMutateResidue doesn't build sidechains and doesn't add hydrogens automatically
oespruce.OEBuildSidechains(target_structure)
oechem.OEPlaceHydrogens(target_structure)
# update residue information
oechem.OEPerceiveResidues(target_structure, oechem.OEPreserveResInfo_All)
return target_structure
