def main(argv=[__name__]):
if len(argv) not in [2, 3]:
oechem.OEThrow.Usage("{} <input protein PDB> [nowrite]".format(argv[0])) # noqa
do_write = True
if len(argv) == 3:
if argv[2] != "nowrite":
oechem.OEThrow.Warning("{} is not a valid option.\n".format(argv[1]))
sys.exit(1)
else:
do_write = False
bio_system_file = argv[1]
bio_system = oechem.OEGraphMol()
input_success = ReadFromPDB(bio_system_file, bio_system)
if not input_success:
oechem.OEThrow.Fatal("Unable to input protein from PDB file.")
# @ <SNIPPET-OEMakeDesignUnits>
design_units = oespruce.OEMakeDesignUnits(bio_system)
# @ </SNIPPET-OEMakeDesignUnits>
if do_write:
base_name = bio_system_file[:-4]
temp_dir = tempfile.mkdtemp()
for i, design_unit in enumerate(design_units):
output_design_unit_file = os.path.join(temp_dir, base_name + "_DU_{}.oedu".format(i)) # noqa
print("\nWriting design unit {} to {}".format(i, output_design_unit_file))
oechem.OEWriteDesignUnit(output_design_unit_file, design_unit)
def prepare_receptor(complex_pdb_filename, prefix):
# Read the receptor and identify design units
from openeye import oespruce, oechem
complex = read_pdb_file(complex_pdb_filename)
print('Identifying design units...')
design_units = list(oespruce.OEMakeDesignUnits(complex))
for i, design_unit in enumerate(design_units):
filename = f'DU_{i}.oedu'
print(f'Writing design unit {i} to {filename}')
oechem.OEWriteDesignUnit(filename, design_unit)
if len(design_units) > 1:
print('More than one design unit found---using first one')
design_unit = design_units[0]
# Prepare the receptor
print('Preparing receptor...')
from openeye import oedocking
protein = oechem.OEGraphMol()
design_unit.GetProtein(protein)
ligand = oechem.OEGraphMol()
design_unit.GetLigand(ligand)
receptor = oechem.OEGraphMol()
oedocking.OEMakeReceptor(receptor, protein, ligand)
oedocking.OEWriteReceptorFile(receptor, f'{prefix}-receptor.oeb.gz')
with oechem.oemolostream(f'{prefix}-protein.pdb') as ofs:
oechem.OEWriteMolecule(ofs, protein)
with oechem.oemolostream(f'{prefix}-ligand.mol2') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
with oechem.oemolostream(f'{prefix}-ligand.pdb') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
with oechem.oemolostream(f'{prefix}-ligand.sdf') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
def make_design_units(
complex: oechem.OEGraphMol, metadata: oespruce.OEStructureMetadata,
options: oespruce.OEMakeDesignUnitOptions
) -> List[oechem.OEDesignUnit]:
dus = list(oespruce.OEMakeDesignUnits(complex, metadata, options))
if len(dus) == 0:
raise RuntimeError('No design units found.')
return dus
def _prepare_structure(
structure: oechem.OEGraphMol,
has_ligand: bool = False,
electron_density: Union[oegrid.OESkewGrid, None] = None,
loop_db: Union[str, None] = None,
ligand_name: Union[str, None] = None,
cap_termini: bool = True,
real_termini: Union[List[int], None] = None,
) -> Union[oechem.OEDesignUnit, None]:
"""
Prepare an OpenEye molecule holding a protein ligand complex for docking.
Parameters
----------
structure: oechem.OEGraphMol
An OpenEye molecule holding a structure with protein and optionally a ligand.
has_ligand: bool
If structure contains a ligand that should be used in design unit generation.
electron_density: oegrid.OESkewGrid
An OpenEye grid holding the electron density.
loop_db: str or None
Path to OpenEye Spruce loop database. You can request a copy at
https://www.eyesopen.com/database-downloads. A testing subset (3TPP) is available
at https://docs.eyesopen.com/toolkits/python/sprucetk/examples_make_design_units.html.
ligand_name: str or None
The name of the ligand located in the binding pocket of interest.
cap_termini: bool
If termini should be capped with ACE and NME.
real_termini: list of int or None
Residue numbers of biologically real termini will not be capped with ACE and NME.
Returns
-------
design_unit: oechem.OEDesignUnit or None
An OpenEye design unit holding the prepared structure with the highest quality among all identified design
units.
"""
def _has_residue_number(atom, residue_numbers=real_termini):
"""Return True if atom matches any given residue number."""
residue = oechem.OEAtomGetResidue(atom)
return any([
True if residue.GetResidueNumber() == residue_number else False
for residue_number in residue_numbers
])
# set design unit options
structure_metadata = oespruce.OEStructureMetadata()
design_unit_options = oespruce.OEMakeDesignUnitOptions()
if cap_termini is False:
design_unit_options.GetPrepOptions().GetBuildOptions().SetCapCTermini(
False)
design_unit_options.GetPrepOptions().GetBuildOptions().SetCapNTermini(
False)
if loop_db is not None:
from pathlib import Path
loop_db = str(Path(loop_db).expanduser().resolve())
design_unit_options.GetPrepOptions().GetBuildOptions(
).GetLoopBuilderOptions().SetLoopDBFilename(loop_db)
# cap all termini but biologically real termini
if real_termini is not None and cap_termini is True:
oespruce.OECapTermini(structure,
oechem.PyAtomPredicate(_has_residue_number))
# already capped, preserve biologically real termini
design_unit_options.GetPrepOptions().GetBuildOptions().SetCapCTermini(
False)
design_unit_options.GetPrepOptions().GetBuildOptions().SetCapNTermini(
False)
# make design units
if has_ligand:
if electron_density is None:
design_units = list(
oespruce.OEMakeDesignUnits(structure, structure_metadata,
design_unit_options))
# filter design units for ligand of interest
if ligand_name is not None:
design_units = [
design_unit for design_unit in design_units
if ligand_name in design_unit.GetTitle()
]
else:
design_units = list(
oespruce.OEMakeDesignUnits(structure, electron_density,
structure_metadata,
design_unit_options))
else:
design_units = list(
oespruce.OEMakeBioDesignUnits(structure, structure_metadata,
design_unit_options))
if len(design_units) >= 1:
design_unit = design_units[0]
else:
# TODO: Returns None if something goes wrong
return None
# fix loop issues and missing OXT
_OEFixLoopIssues(design_unit)
_OEFixMissingOXT(design_unit)
return design_unit
def prepare_receptor(complex_pdb_filename, output_basepath, dimer=False):
"""
Parameters
----------
complex_pdb_filename : str
The complex PDB file to read in
output_basepath : str
Base path for output
dimer : bool, optional, default=False
If True, generate the dimer as the biological unit
"""
import os
basepath, filename = os.path.split(complex_pdb_filename)
prefix, extension = os.path.splitext(filename)
prefix = os.path.join(output_basepath, prefix)
# Check if receptor already exists
receptor_filename = f'{prefix}-receptor.oeb.gz'
thiolate_receptor_filename = f'{prefix}-receptor-thiolate.oeb.gz'
if os.path.exists(receptor_filename) and os.path.exists(
thiolate_receptor_filename):
return
# Read in PDB file
pdbfile_lines = [
line for line in open(complex_pdb_filename, 'r') if 'UNK' not in line
]
# If monomer is specified, drop crystal symmetry lines
if not dimer:
pdbfile_lines = [
line for line in pdbfile_lines if 'REMARK 350' not in line
]
# Reconstruct PDBFile contents
pdbfile_contents = ''.join(pdbfile_lines)
# Read the receptor and identify design units
from openeye import oespruce, oechem
from tempfile import NamedTemporaryFile
with NamedTemporaryFile(delete=False, mode='wt', suffix='.pdb') as pdbfile:
pdbfile.write(pdbfile_contents)
pdbfile.close()
complex = read_pdb_file(pdbfile.name)
# TODO: Clean up
#print('Identifying design units...')
design_units = list(oespruce.OEMakeDesignUnits(complex))
if len(design_units) == 1:
design_unit = design_units[0]
elif len(design_units) > 1:
#print('More than one design unit found---using first one')
design_unit = design_units[0]
elif len(design_units) == 0:
raise Exception(f' * No design units found for {complex_pdb_filename}')
# Prepare the receptor
#print('Preparing receptor...')
from openeye import oedocking
protein = oechem.OEGraphMol()
design_unit.GetProtein(protein)
ligand = oechem.OEGraphMol()
design_unit.GetLigand(ligand)
receptor = oechem.OEGraphMol()
oedocking.OEMakeReceptor(receptor, protein, ligand)
oedocking.OEWriteReceptorFile(receptor, receptor_filename)
with oechem.oemolostream(f'{prefix}-protein.pdb') as ofs:
oechem.OEWriteMolecule(ofs, protein)
with oechem.oemolostream(f'{prefix}-ligand.mol2') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
with oechem.oemolostream(f'{prefix}-ligand.pdb') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
with oechem.oemolostream(f'{prefix}-ligand.sdf') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
# Filter out UNK from PDB files (which have covalent adducts)
pdbfile_lines = [
line for line in open(f'{prefix}-protein.pdb', 'r')
if 'UNK' not in line
]
with open(f'{prefix}-protein.pdb', 'wt') as outfile:
outfile.write(''.join(pdbfile_lines))
# Adjust protonation state of CYS145 to generate thiolate form
#print('Deprotonating CYS145...')
pred = oechem.OEAtomMatchResidue(["CYS:145: :A"])
for atom in protein.GetAtoms(pred):
if oechem.OEGetPDBAtomIndex(atom) == oechem.OEPDBAtomName_SG:
oechem.OESuppressHydrogens(atom)
atom.SetFormalCharge(-1)
atom.SetImplicitHCount(0)
# Adjust protonation states
#print('Re-optimizing hydrogen positions...')
place_hydrogens_opts = oechem.OEPlaceHydrogensOptions()
place_hydrogens_opts.SetBypassPredicate(pred)
protonate_opts = oespruce.OEProtonateDesignUnitOptions(
place_hydrogens_opts)
success = oespruce.OEProtonateDesignUnit(design_unit, protonate_opts)
design_unit.GetProtein(protein)
# Old hacky way to adjust protonation states
#opts = oechem.OEPlaceHydrogensOptions()
#opts.SetBypassPredicate(pred)
#describe = oechem.OEPlaceHydrogensDetails()
#success = oechem.OEPlaceHydrogens(protein, describe, opts)
#if success:
# oechem.OEUpdateDesignUnit(design_unit, protein, oechem.OEDesignUnitComponents_Protein)
# Write thiolate form of receptor
receptor = oechem.OEGraphMol()
oedocking.OEMakeReceptor(receptor, protein, ligand)
oedocking.OEWriteReceptorFile(receptor, thiolate_receptor_filename)
with oechem.oemolostream(f'{prefix}-protein-thiolate.pdb') as ofs:
oechem.OEWriteMolecule(ofs, protein)
# Filter out UNK from PDB files (which have covalent adducts)
pdbfile_lines = [
line for line in open(f'{prefix}-protein-thiolate.pdb', 'r')
if 'UNK' not in line
]
with open(f'{prefix}-protein-thiolate.pdb', 'wt') as outfile:
outfile.write(''.join(pdbfile_lines))
def prepare_receptor(complex_pdb_filename,
output_basepath,
dimer=False,
retain_water=False):
"""
Parameters
----------
complex_pdb_filename : str
The complex PDB file to read in
output_basepath : str
Base path for output
dimer : bool, optional, default=False
If True, generate the dimer as the biological unit
retain_water : bool, optional, default=False
If True, will retain waters
"""
# Check whether this is a diamond SARS-CoV-2 Mpro structure or not
import re
is_diamond_structure = (re.search('-x\d+_', complex_pdb_filename)
is not None)
import os
basepath, filename = os.path.split(complex_pdb_filename)
prefix, extension = os.path.splitext(filename)
prefix = os.path.join(output_basepath, prefix)
# Check if receptor already exists
receptor_filename = f'{prefix}-receptor.oeb.gz'
thiolate_receptor_filename = f'{prefix}-receptor-thiolate.oeb.gz'
if os.path.exists(receptor_filename) and os.path.exists(
thiolate_receptor_filename):
return
# Read in PDB file, skipping UNK atoms (left over from processing covalent ligands)
pdbfile_lines = [
line for line in open(complex_pdb_filename, 'r') if 'UNK' not in line
]
# Check if biological symmetry header is present
has_biological_symmetry_header = False
for line in pdbfile_lines:
if 'REMARK 350' in line:
has_biological_symmetry_header = True
break
# Prepend REMARK 350 (biological symmetry) header lines for Mpro (from 5RGG) if not present
if is_diamond_structure and (not has_biological_symmetry_header):
pdbfile_lines = [
line + '\n' for line in BIOLOGICAL_SYMMETRY_HEADER.split('\n')
] + pdbfile_lines
# If monomer is specified, drop crystal symmetry lines
if not dimer:
pdbfile_lines = [
line for line in pdbfile_lines if 'REMARK 350' not in line
]
# Filter out waters
if not retain_water:
pdbfile_lines = [line for line in pdbfile_lines if 'HOH' not in line]
# Filter out LINK records to covalent inhibitors so we can model non-covalent complex
pdbfile_lines = [line for line in pdbfile_lines if 'LINK' not in line]
# Reconstruct PDBFile contents
pdbfile_contents = ''.join(pdbfile_lines)
# Append SEQRES to all structures if they do not have it
seqres = """\
SEQRES 1 A 306 SER GLY PHE ARG LYS MET ALA PHE PRO SER GLY LYS VAL
SEQRES 2 A 306 GLU GLY CYS MET VAL GLN VAL THR CYS GLY THR THR THR
SEQRES 3 A 306 LEU ASN GLY LEU TRP LEU ASP ASP VAL VAL TYR CYS PRO
SEQRES 4 A 306 ARG HIS VAL ILE CYS THR SER GLU ASP MET LEU ASN PRO
SEQRES 5 A 306 ASN TYR GLU ASP LEU LEU ILE ARG LYS SER ASN HIS ASN
SEQRES 6 A 306 PHE LEU VAL GLN ALA GLY ASN VAL GLN LEU ARG VAL ILE
SEQRES 7 A 306 GLY HIS SER MET GLN ASN CYS VAL LEU LYS LEU LYS VAL
SEQRES 8 A 306 ASP THR ALA ASN PRO LYS THR PRO LYS TYR LYS PHE VAL
SEQRES 9 A 306 ARG ILE GLN PRO GLY GLN THR PHE SER VAL LEU ALA CYS
SEQRES 10 A 306 TYR ASN GLY SER PRO SER GLY VAL TYR GLN CYS ALA MET
SEQRES 11 A 306 ARG PRO ASN PHE THR ILE LYS GLY SER PHE LEU ASN GLY
SEQRES 12 A 306 SER CYS GLY SER VAL GLY PHE ASN ILE ASP TYR ASP CYS
SEQRES 13 A 306 VAL SER PHE CYS TYR MET HIS HIS MET GLU LEU PRO THR
SEQRES 14 A 306 GLY VAL HIS ALA GLY THR ASP LEU GLU GLY ASN PHE TYR
SEQRES 15 A 306 GLY PRO PHE VAL ASP ARG GLN THR ALA GLN ALA ALA GLY
SEQRES 16 A 306 THR ASP THR THR ILE THR VAL ASN VAL LEU ALA TRP LEU
SEQRES 17 A 306 TYR ALA ALA VAL ILE ASN GLY ASP ARG TRP PHE LEU ASN
SEQRES 18 A 306 ARG PHE THR THR THR LEU ASN ASP PHE ASN LEU VAL ALA
SEQRES 19 A 306 MET LYS TYR ASN TYR GLU PRO LEU THR GLN ASP HIS VAL
SEQRES 20 A 306 ASP ILE LEU GLY PRO LEU SER ALA GLN THR GLY ILE ALA
SEQRES 21 A 306 VAL LEU ASP MET CYS ALA SER LEU LYS GLU LEU LEU GLN
SEQRES 22 A 306 ASN GLY MET ASN GLY ARG THR ILE LEU GLY SER ALA LEU
SEQRES 23 A 306 LEU GLU ASP GLU PHE THR PRO PHE ASP VAL VAL ARG GLN
SEQRES 24 A 306 CYS SER GLY VAL THR PHE GLN
"""
has_seqres = 'SEQRES' in pdbfile_contents
if not has_seqres:
#print('Adding SEQRES')
pdbfile_contents = seqres + pdbfile_contents
# Read the receptor and identify design units
from openeye import oespruce, oechem
from tempfile import NamedTemporaryFile
with NamedTemporaryFile(delete=False, mode='wt', suffix='.pdb') as pdbfile:
pdbfile.write(pdbfile_contents)
pdbfile.close()
complex = read_pdb_file(pdbfile.name)
# TODO: Clean up
# Strip protons from structure to allow SpruceTK to add these back
# See: 6wnp, 6wtj, 6wtk, 6xb2, 6xqs, 6xqt, 6xqu, 6m2n
#print('Suppressing hydrogens')
#print(f' Initial: {sum([1 for atom in complex.GetAtoms()])} atoms')
for atom in complex.GetAtoms():
if atom.GetAtomicNum() > 1:
oechem.OESuppressHydrogens(atom)
#print(f' Final: {sum([1 for atom in complex.GetAtoms()])} atoms')
# Delete and rebuild C-terminal residue because Spruce causes issues with this
# See: 6m2n 6lze
#print('Deleting C-terminal residue O')
pred = oechem.OEIsCTerminalAtom()
for atom in complex.GetAtoms():
if pred(atom):
for nbor in atom.GetAtoms():
if oechem.OEGetPDBAtomIndex(nbor) == oechem.OEPDBAtomName_O:
complex.DeleteAtom(nbor)
#pred = oechem.OEAtomMatchResidue(["GLN:306:.*:.*:.*"])
#for atom in complex.GetAtoms(pred):
# if oechem.OEGetPDBAtomIndex(atom) == oechem.OEPDBAtomName_O:
# print('Deleting O')
# complex.DeleteAtom(atom)
#het = oespruce.OEHeterogenMetadata()
#het.SetTitle("LIG") # real ligand 3 letter code
#het.SetID("CovMoonShot1234") # in case you have corporate IDs
#het.SetType(oespruce.OEHeterogenType_Ligand)
# mdata.AddHeterogenMetadata(het)
#print('Identifying design units...')
# Produce zero design units if we fail to protonate
# Log warnings
errfs = oechem.oeosstream(
) # create a stream that writes internally to a stream
oechem.OEThrow.SetOutputStream(errfs)
oechem.OEThrow.Clear()
oechem.OEThrow.SetLevel(
oechem.OEErrorLevel_Verbose) # capture verbose error output
opts = oespruce.OEMakeDesignUnitOptions()
#print(f'ligand atoms: min {opts.GetSplitOptions().GetMinLigAtoms()}, max {opts.GetSplitOptions().GetMaxLigAtoms()}')
opts.GetSplitOptions().SetMinLigAtoms(
7) # minimum fragment size (in heavy atoms)
mdata = oespruce.OEStructureMetadata()
opts.GetPrepOptions().SetStrictProtonationMode(True)
# Both N- and C-termini should be zwitterionic
# Mpro cleaves its own N- and C-termini
# See https://www.pnas.org/content/113/46/12997
opts.GetPrepOptions().GetBuildOptions().SetCapNTermini(False)
opts.GetPrepOptions().GetBuildOptions().SetCapCTermini(False)
# Don't allow truncation of termini, since force fields don't have parameters for this
opts.GetPrepOptions().GetBuildOptions().GetCapBuilderOptions(
).SetAllowTruncate(False)
# Build loops and sidechains
opts.GetPrepOptions().GetBuildOptions().SetBuildLoops(True)
opts.GetPrepOptions().GetBuildOptions().SetBuildSidechains(True)
# Don't flip Gln189
#pred = oechem.OEAtomMatchResidue(["GLN:189: :A"])
pred = oechem.OEAtomMatchResidue(["GLN:189:.*:.*:.*"])
protonate_opts = opts.GetPrepOptions().GetProtonateOptions()
place_hydrogens_opts = protonate_opts.GetPlaceHydrogensOptions()
#place_hydrogens_opts.SetBypassPredicate(pred)
place_hydrogens_opts.SetNoFlipPredicate(pred)
#protonate_opts = oespruce.OEProtonateDesignUnitOptions(place_hydrogens_opts)
#opts.GetPrepOptions().SetProtonateOptions(protonate_options);
# Make design units
design_units = list(oespruce.OEMakeDesignUnits(complex, mdata, opts))
# Restore error stream
oechem.OEThrow.SetOutputStream(oechem.oeerr)
# Capture the warnings to a string
warnings = errfs.str().decode("utf-8")
if len(design_units) >= 1:
design_unit = design_units[0]
print('')
print('')
print(f'{complex_pdb_filename} : SUCCESS')
print(warnings)
elif len(design_units) == 0:
print('')
print('')
print(f'{complex_pdb_filename} : FAILURE')
print(warnings)
msg = f'No design units found for {complex_pdb_filename}\n'
msg += warnings
msg += '\n'
raise Exception(msg)
# Prepare the receptor
#print('Preparing receptor...')
from openeye import oedocking
protein = oechem.OEGraphMol()
design_unit.GetProtein(protein)
ligand = oechem.OEGraphMol()
design_unit.GetLigand(ligand)
# Create receptor and other files
receptor = oechem.OEGraphMol()
oedocking.OEMakeReceptor(receptor, protein, ligand)
oedocking.OEWriteReceptorFile(receptor, receptor_filename)
with oechem.oemolostream(f'{prefix}-protein.pdb') as ofs:
oechem.OEWriteMolecule(ofs, protein)
with oechem.oemolostream(f'{prefix}-ligand.mol2') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
with oechem.oemolostream(f'{prefix}-ligand.pdb') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
with oechem.oemolostream(f'{prefix}-ligand.sdf') as ofs:
oechem.OEWriteMolecule(ofs, ligand)
# Filter out UNK from PDB files (which have covalent adducts)
pdbfile_lines = [
line for line in open(f'{prefix}-protein.pdb', 'r')
if 'UNK' not in line
]
with open(f'{prefix}-protein.pdb', 'wt') as outfile:
outfile.write(''.join(pdbfile_lines))
# Adjust protonation state of CYS145 to generate thiolate form
#print('Deprotonating CYS145...') # DEBUG
#pred = oechem.OEAtomMatchResidue(["CYS:145: :A"])
pred = oechem.OEAtomMatchResidue(["CYS:145:.*:.*:.*"])
place_hydrogens_opts.SetBypassPredicate(pred)
for atom in protein.GetAtoms(pred):
if oechem.OEGetPDBAtomIndex(atom) == oechem.OEPDBAtomName_SG:
#print('Modifying CYS 145 SG')
oechem.OESuppressHydrogens(atom)
atom.SetFormalCharge(-1)
atom.SetImplicitHCount(0)
#print('Protonating HIS41...') # DEBUG
#pred = oechem.OEAtomMatchResidue(["HIS:41: :A"])
pred = oechem.OEAtomMatchResidue(["HIS:41:.*:.*:.*"])
place_hydrogens_opts.SetBypassPredicate(pred)
for atom in protein.GetAtoms(pred):
if oechem.OEGetPDBAtomIndex(atom) == oechem.OEPDBAtomName_ND1:
#print('Protonating HIS 41 ND1')
oechem.OESuppressHydrogens(atom) # strip hydrogens from residue
atom.SetFormalCharge(+1)
atom.SetImplicitHCount(1)
# Update the design unit with the modified formal charge for CYS 145 SG
oechem.OEUpdateDesignUnit(design_unit, protein,
oechem.OEDesignUnitComponents_Protein)
# Don't flip Gln189
#pred = oechem.OEAtomMatchResidue(["GLN:189: :A"])
#protonate_opts = opts.GetPrepOptions().GetProtonateOptions();
#place_hydrogens_opts = protonate_opts.GetPlaceHydrogensOptions()
#place_hydrogens_opts.SetNoFlipPredicate(pred)
# Adjust protonation states
#print('Re-optimizing hydrogen positions...') # DEBUG
#place_hydrogens_opts = oechem.OEPlaceHydrogensOptions()
#place_hydrogens_opts.SetBypassPredicate(pred)
#protonate_opts = oespruce.OEProtonateDesignUnitOptions(place_hydrogens_opts)
success = oespruce.OEProtonateDesignUnit(design_unit, protonate_opts)
design_unit.GetProtein(protein)
# Write thiolate form of receptor
receptor = oechem.OEGraphMol()
oedocking.OEMakeReceptor(receptor, protein, ligand)
oedocking.OEWriteReceptorFile(receptor, thiolate_receptor_filename)
with oechem.oemolostream(f'{prefix}-protein-thiolate.pdb') as ofs:
oechem.OEWriteMolecule(ofs, protein)
# Filter out UNK from PDB files (which have covalent adducts)
pdbfile_lines = [
line for line in open(f'{prefix}-protein-thiolate.pdb', 'r')
if 'UNK' not in line
]
with open(f'{prefix}-protein-thiolate.pdb', 'wt') as outfile:
outfile.write(''.join(pdbfile_lines))
def prepare_complex(protein_ligand_complex,
electron_density=None,
loop_db=None,
protein_save_path=None,
ligand_save_path=None):
"""
Prepare a OEChem molecule holding a protein ligand complex for docking.
Parameters
----------
protein_ligand_complex: oechem.OEGraphMol
An oechem.OEGraphMol object holding a structure with protein and ligand.
electron_density: oegrid.OESkewGrid
An oegrid.OESkewGrid object holding the electron density.
loop_db: str
File path for OpenEye Spruce loop database.
protein_save_path: str
File path for saving prepared protein. If protein_save_path is not provided, protein will not be saved.
ligand_save_path: str
File path for saving prepared ligand. If ligand_save_path is not provided, ligand will not be saved.
Returns
-------
protein: oechem.OEGraphMol
An oechem.OEGraphMol object holding a prepared protein structure.
ligand: oechem.OEGraphMol
An oechem.OEGraphMol object holding a prepared ligand structure.
"""
# Standard libraries
import pathlib
# External libraries
from openeye import oechem, oespruce
# create design units
structure_metadata = oespruce.OEStructureMetadata()
design_unit_options = oespruce.OEMakeDesignUnitOptions()
if loop_db is not None:
design_unit_options.GetPrepOptions().GetBuildOptions(
).GetLoopBuilderOptions().SetLoopDBFilename(loop_db)
if electron_density is None:
design_units = list(
oespruce.OEMakeDesignUnits(protein_ligand_complex,
structure_metadata,
design_unit_options))
else:
design_units = list(
oespruce.OEMakeDesignUnits(protein_ligand_complex,
electron_density, structure_metadata,
design_unit_options))
if len(design_units) == 1:
design_unit = design_units[0]
elif len(design_units) > 1:
print('More than one design unit found---using first one')
design_unit = design_units[0]
else:
print('No design units found')
return None, None
# get protein
protein = oechem.OEGraphMol()
design_unit.GetProtein(protein)
# get ligand
ligand = oechem.OEGraphMol()
design_unit.GetLigand(ligand)
# save protein
if protein_save_path is not None:
with oechem.oemolostream(
str(pathlib.Path(protein_save_path).absolute())) as ofs:
oechem.OEWriteMolecule(ofs, protein)
# save ligand
if ligand_save_path is not None:
with oechem.oemolostream(str(
pathlib.Path(ligand_save_path).absolute())) as ofs:
oechem.OEWriteMolecule(ofs, ligand)
return protein, ligand