def create_feature_db(mol_files, outdir, dbname="test"):
# because CM is installed in a non-standard location
f_defs = os.path.join(os.path.dirname(os.path.dirname(io.csd_directory())),
"CSD_CrossMiner/feature_definitions")
Pharmacophore.read_feature_definitions(f_defs)
sdbs = []
for mol_file in mol_files:
# DatabaseInfo is a named tupled (file name, num_strucs, colour)
mol_struc = Pharmacophore.FeatureDatabase.DatabaseInfo(
mol_file, 0, Colour(0, 255, 0, 255))
# Create structure databases
mol_sqlx = os.path.join(
outdir,
os.path.basename(mol_file).replace('.mol2', '.csdsqlx'))
if not os.path.exists(outdir):
os.mkdir(outdir)
mol_sdb = Pharmacophore.FeatureDatabase.Creator.StructureDatabase(
mol_struc,
use_crystal_symmetry=False,
structure_database_path=mol_sqlx)
sdbs.append(mol_sdb)
# Create Feature database
settings = Pharmacophore.FeatureDatabase.Creator.Settings(
feature_definition_directory=f_defs, n_threads=6)
creator = Pharmacophore.FeatureDatabase.Creator(settings=settings)
db = creator.create(sdbs)
db.write(os.path.join(outdir, f"{dbname}.feat"))
def __init__(self, **kw):
settings = kw.get('settings')
if settings is None:
settings = self.Settings()
self.settings = settings
main_dir = os.environ.get('MAINDIR')
if main_dir:
if sys.platform == 'win32':
self.settings._superstar_executable = 'superstar_app.exe'
else:
self.settings._superstar_executable = ' '.join([
os.path.join(os.environ['MAINDIR'], 'run.sh'),
'superstar_app.x'
])
self.settings._superstar_env = dict()
else:
if sys.platform == 'win32':
base = os.path.dirname(io.csd_directory())
merc = glob.glob(os.path.join(base, 'mercury*'))
if len(merc):
merc = merc[0]
self.settings._superstar_executable = os.path.join(
merc, 'superstar_app.exe')
elif sys.platform == 'darwin':
self.settings._superstar_executable = os.path.join(
os.path.dirname(io.csd_directory()), 'mercury.app',
'Contents', 'MacOS', 'superstar')
else:
self.settings._superstar_executable = os.path.join(
os.path.dirname(io.csd_directory()), 'bin', 'superstar')
self.settings._superstar_env = dict(
SUPERSTAR_ISODIR=str(
os.path.join(os.path.dirname(io.csd_directory()),
'isostar_files', 'istr')),
SUPERSTAR_ROOT=str(
os.path.join(os.path.dirname(io.csd_directory()),
"Mercury")))
self.settings.working_directory = utilities._test_output_dir()
def _set_environment_variables():
"""
private method
sets up superstar environment variables
:return: superstar executable (str), superstar env(str)
"""
base = csd_directory()
main_dir = environ.get('MAINDIR')
if main_dir:
if sys.platform == 'win32':
superstar_executable = 'superstar_app.exe'
else:
superstar_executable = ' '.join([
join(environ['MAINDIR'], 'run.sh'), 'superstar_app.x'
])
superstar_env = dict()
else:
if sys.platform == 'win32':
base = dirname(base)
merc = glob.glob(join(base, 'mercury*'))
if type(merc) is list:
try:
merc = merc[0]
except IndexError:
raise IndexError(
"No mercury path found, check API version")
superstar_executable = join(merc, 'superstar_app.exe')
if not isfile(superstar_executable):
superstar_executable = join(merc, 'superstar.exe')
if not isfile(superstar_executable):
raise IOError("superstar executable not found")
superstar_env = dict(SUPERSTAR_ISODIR=str(
join(base, 'isostar_files', 'istr')),
SUPERSTAR_ROOT=str(
join(base, "Mercury")))
elif sys.platform == 'darwin':
print("OS X not supported")
else:
base = dirname(base)
superstar_executable = join(base, 'bin', 'superstar')
superstar_env = dict(SUPERSTAR_ISODIR=str(
join(base, 'isostar_files', 'istr')),
SUPERSTAR_ROOT=str(base))
return superstar_executable, superstar_env
def setUp(self):
self.parent_dir = "testdata/pharmacophore_extension/PharmacophoreModel"
self.fnames = [
"1dmt_ligand.cm", "1r1h_ligand.cm", "1r1j_ligand.cm",
"1y8j_ligand.cm"
]
self.pharmacophores = [
PharmacophoreModel.from_file(os.path.join(self.parent_dir, f))
for f in self.fnames
]
self.cm_dir = os.path.dirname(os.path.dirname(io.csd_directory()))
Pharmacophore.read_feature_definitions(
os.path.join(self.cm_dir, "CSD_CrossMiner/feature_definitions"))
def __init__(self):
super().__init__()
self.cm_dir = os.path.dirname(os.path.dirname(csd_directory()))
Pharmacophore.read_feature_definitions(directory=os.path.join(
self.cm_dir, "CSD_CrossMiner/feature_definitions"))
self.__feature_options = {
k: v
for k, v in Pharmacophore.feature_definitions.items()
}
assert len(self.__feature_options) > 1
self.__feature_definitions = self.__feature_options
self.tmp = tempfile.mkdtemp()
self.__identifier = None
self.__ligands = None
self.__protein = None
self.__detected_features = None
self.__feature_point_grids = None
def __init__(self, features=None, _motif_pharmacophore=None):
super().__init__(features=features,
_motif_pharmacophore=_motif_pharmacophore)
self.cm_dir = os.path.dirname(os.path.dirname(csd_directory()))
feat_db = os.environ.get(
"CCDC_CROSSMINER_FEATURE_DEFINITIONS",
os.path.join(self.cm_dir, "../CSD_CrossMiner/feature_definitions"))
Pharmacophore.read_feature_definitions(directory=feat_db)
self.__feature_options = {
k: v
for k, v in Pharmacophore.feature_definitions.items()
}
assert len(self.__feature_options) > 1
self.__feature_definitions = self.__feature_options
self.tmp = tempfile.mkdtemp()
self.__identifier = None
self.__ligands = None
self.__protein = None
self.__detected_features = None
self.__feature_point_grids = None
from ccdc.pharmacophore import Pharmacophore
from ccdc import io
import os
from shutil import copyfile
from ccdc.utilities import Colour, Timer
if __name__ == "__main__":
outdir = "/home/pcurran/github_packages/pharmacophores/testdata/search/feat_db"
f_defs = os.path.join(os.path.dirname(os.path.dirname(io.csd_directory())),
"CSD_CrossMiner/feature_definitions")
Pharmacophore.read_feature_definitions(f_defs)
base = "/local/pcurran/patel/CDK2/screening_files/conformers"
mol_files = [
os.path.join(base, f) for f in
["actives_final_chunk0_conf.mol2", "decoys_final_chunk0_conf.mol2"]
]
sdbs = []
for mol_file in mol_files:
# DatabaseInfo is a named tupled (file name, num_strucs, colour)
mol_struc = Pharmacophore.FeatureDatabase.DatabaseInfo(
mol_file, 0, Colour(0, 255, 0, 255))
# Create structure databases
mol_sqlx = os.path.join(
outdir,
os.path.basename(mol_file).replace('.mol2', '.csdsqlx'))
if not os.path.exists(outdir):
os.mkdir(outdir)
def search(self,
queryTargetId,
queryTargetPath,
resultPath,
normalizeFlag=True,
maxHits=50,
searchType="similarity",
suppressMetals=False):
"""Search the CCDC database for similar or substructure matches for the input query molecule.
Args:
queryTargetId (str): query identifier
queryTargetPath (str): path to the query molfile (mol, sdf, mol2)
resultPath (str): output path to match results
normalizeFlag (bool, optional): do standard perceptions on matching molecules. Defaults to True.
maxHits (int, optional): maximum number of matches to return. Defaults to 50.
searchType (str, optional): search mode (substructure, similarity). Defaults to "similarity".
suppressMetals (bool, optional): filter structures containing metals. Defaults to False.
Returns:
(int): number of matches
"""
mU = MarshalUtil()
logger.info("Start search for target %s path %s result path %s",
queryTargetId, queryTargetPath, resultPath)
#
summaryList = []
#
targetDirPath = os.path.dirname(queryTargetPath)
cifTargetPath = os.path.join(targetDirPath, queryTargetId + ".cif")
#
targetStructures = EntryReader(queryTargetPath)
dirPath = os.path.join(resultPath, queryTargetId)
numHits = 0
for ii, e in enumerate(targetStructures, 1):
numHits = 0
startTime = time.time()
targetMol = e.molecule
if normalizeFlag:
targetMol.assign_bond_types(which="unknown")
targetMol.standardise_aromatic_bonds()
targetMol.standardise_delocalised_bonds()
#
logger.info("(%d) begin %s search - query id %s", ii, searchType,
queryTargetId)
if searchType == "similarity":
hits = self.__similaritySearch(targetMol,
suppressMetals=suppressMetals)
elif searchType == "substructure":
hits = self.__moleculeSubstructureSearch(
targetMol, suppressMetals=suppressMetals)
else:
hits = []
logger.info("(%d) completed search query id %s in %.3f seconds",
ii, queryTargetId,
time.time() - startTime)
if hits:
numHits += len(hits)
logger.info("(%d) search for %s matched %d: %r", ii,
queryTargetId, numHits,
[targetHit.identifier for targetHit in hits])
#
for targetHit in hits[:maxHits]:
#
hI = CcdcMatchIndexInst()
hI.setCsdVersion(csd_version())
hI.setCsdDirectory(csd_directory())
hI.setTargetId(queryTargetId)
hI.setTargetPath(queryTargetPath)
if mU.exists(cifTargetPath):
hI.setTargetCcPath(cifTargetPath)
hI.setIdentifier(targetHit.identifier)
hI.setMatchType(searchType)
try:
hI.setRFactor(targetHit.entry.r_factor)
hI.setChemicalName(targetHit.entry.chemical_name)
hI.setTemperature(targetHit.entry.temperature)
hI.setRadiationSource(targetHit.entry.radiation_source)
hI.setHasDisorder("N")
cit = targetHit.entry.publication
if cit.doi is not None:
hI.setCitationDOI(cit.doi)
if searchType == "similarity":
hI.setSimilarityScore(targetHit.similarity)
elif searchType == "substructure":
hI.setMatchedAtomLength(
len(targetHit.match_atoms()))
except Exception as e:
logger.exception("Failing with %s", str(e))
#
#
mU.mkdir(dirPath)
mol2L = []
if searchType == "substructure":
for jj, mc in enumerate(targetHit.match_components(),
1):
fp = os.path.join(
dirPath, queryTargetId + "_" +
targetHit.identifier + "_%03d" % jj + ".mol2")
mol2L.append(fp)
with MoleculeWriter(fp) as ofh:
ofh.write(mc)
# Replace the title line
with open(fp) as fin:
lines = fin.readlines()
lines[1] = lines[1].replace(
"00", targetHit.identifier)
#
with open(fp, "w") as fout:
fout.write("".join(lines))
#
fp = os.path.join(
dirPath, queryTargetId + "_" +
targetHit.identifier + "_%03d" % jj + ".sdf")
with MoleculeWriter(fp) as ofh:
ofh.write(mc)
# Replace the title line
with open(fp) as fin:
lines = fin.readlines()
lines[0] = lines[0].replace(
"00", targetHit.identifier)
#
with open(fp, "w") as fout:
fout.write("".join(lines))
#
# Check for multiple generated result files -
#
for jj, fp in enumerate(mol2L, 1):
logger.debug("(%d) adding component fp %s", jj, fp)
hI.setMatchNumber(jj)
hI.setMol2Path(fp)
tt = fp[:-4] + "sdf"
hI.setMolPath(tt)
summaryList.append(copy.deepcopy(hI.get()))
#
else:
hI.setMatchNumber(1)
summaryList.append(copy.deepcopy(hI.get()))
else:
logger.info("(%d) search for %s returns no matches", ii,
targetMol.identifier)
hits = None
#
if numHits > 0:
mU.mkdir(dirPath)
fp = os.path.join(dirPath, queryTargetId + "-index.json")
cmI = CcdcMatchIndex(indexFilePath=fp, verbose=self.__verbose)
cmI.load(summaryList)
cmI.writeIndex()
return numHits
def delete_solvents(self, list_solvent_names=None):
"""删除晶体中的溶剂，若没有指定溶剂列表，则默认为CCDC数据库自带的溶剂列表
:param list_solvent_names: 溶剂名称构成的列表，type：list or tuple
:return: None
"""
# CSD数据库的溶剂所在的路径
solvent_file = os.path.join(os.path.dirname(io.csd_directory()),
'Mercury', 'molecular_libraries',
'ccdc_solvents')
# 若没指定需要去除的溶剂列表，则会将CSD数据库中指定的74个溶剂都考虑进去。以下代码得到溶剂的smiles字符串
if not list_solvent_names:
if os.path.isdir(solvent_file):
solvent_smiles = [
io.MoleculeReader(f)[0].smiles
for f in glob.glob(os.path.join(solvent_file, '*.mol2'))
]
else:
raise FileExistsError('路径不存在！')
else:
if os.path.isdir(solvent_file):
solvent_smiles = [
io.MoleculeReader(
os.path.join(solvent_file, solvent + '.mol2')[0].smiles
for solvent in list_solvent_names)
]
else:
raise FileExistsError('路径不存在！')
# 去除溶剂
list_crystals_remove_solvents = []
p_bar = tqdm(self.entry_reader)
for entry in p_bar:
try:
if entry.has_3d_structure:
# Ensure labels are unique
mol = entry.molecule
mol.normalise_labels()
# Use a copy
clone = mol.copy()
# Remove all bonds containing a metal atom
clone.remove_bonds(b for b in clone.bonds
if any(a.is_metal for a in b.atoms))
# Work out which components to remove
to_remove = [
c for c in clone.components
if not self.has_metal(c) and (not self.is_multidentate(
c, mol) or self.is_solvent(c, solvent_smiles))
]
# Remove the atoms of selected components
mol.remove_atoms(
mol.atom(a.label) for c in to_remove for a in c.atoms)
# Write the CIF
entry.crystal.molecule = mol
list_crystals_remove_solvents.append(entry)
else:
list_crystals_remove_solvents.append(entry)
except BaseException:
list_crystals_remove_solvents.append(entry)
p_bar.set_description('正在去除溶剂：')
self.entry_reader = list_crystals_remove_solvents
return None
pm = [
PharmacophoreModel.from_file(os.path.join(wrk_dir, f)) for f in fnames
]
feats = create_consensus(pm, cutoff=1)
out = PharmacophoreModel()
out.detected_features = feats
for feat in feats:
out.add_feature(feat)
out.pymol_visulisation(
"/home/pcurran/github_packages/pharmacophores/testdata/concensus")
if __name__ == "__main__":
cm_dir = os.path.dirname(os.path.dirname(io.csd_directory()))
Pharmacophore.read_feature_definitions(
os.path.join(cm_dir, "CSD_CrossMiner/feature_definitions"))
wrkdir = "/home/pcurran/github_packages/pharmacophores/testdata/alignment"
paths = [
"1AQ1_aligned.pdb", "1B38_aligned.pdb", "1B39_aligned.pdb",
"1CKP_aligned.pdb"
]
hetids = ["STU", "ATP", "ATP", "PVB"]
chains = ["A", "A", "A", "A"]
for path, het, chain in zip(paths, hetids, chains):
create_pharmacophore(path, het, chain, out_dir=wrkdir)
wrk_dir = "/home/pcurran/github_packages/pharmacophores/testdata/alignment"