def read_fragment(self, name=None, fmt=None, handle=None,
zero_to_origin=True):
"""Read a molecular structure from a file. Guess at the file type from
extension if caller does not supply explicit fmt. If file handle is
provided, read data from it. Otherwise open file name for reading.
:param name: file to open
:type name : str
:param fmt: optional OpenBabel format code e.g. "xyz"
:type fmt : str
:param zero_to_origin: translate geometry to put atom 0 at origin
:type zero_to_origin : bool
:return: molecular fragment
:rtype : Fragment
"""
if not fmt:
try:
fmt = name.rsplit(".", 1)[-1]
except (IndexError, AttributeError):
msg = "No fmt given for {0} and unable to guess from file extension".format(repr(name))
if handle is None:
molecule = pybel.readfile(fmt, name).next()
else:
data = handle.read()
molecule = pybel.readstring(fmt, data)
fragment = Fragment(molecule)
if zero_to_origin:
fragment.set_zero_to_origin()
return fragment
def prepare(target):
for typ in ('actives', 'decoys'):
url = URL % (target.lower(), typ)
d = os.path.join("Targets_test", target, typ)
if not os.path.isdir(d):
os.makedirs(d)
molfn = os.path.join(d, os.path.basename(url))
if not os.path.isfile(molfn):
print "Downloading %s" % url
i = urllib2.urlopen(url)
o = open(molfn, 'wb')
o.write(i.read())
i.close()
o.close()
#########Only 7 RB #########
if not os.path.isfile(molfn.replace('.sdf.gz', '_7_RB.sdf')):
o = pybel.Outputfile('sdf', molfn.replace('.sdf.gz', '_7_RB.sdf'))
n = 0
for mol in pybel.readfile('sdf', molfn):
if mol.OBMol.NumRotors() <= 7:
o.write(mol)
n += 1
o.close()
if not n:
try:
os.remove(molfn.replace('.sdf.gz', '_7_RB.sdf'))
except:
pass
#########################
clean(validate(molfn, typ))
return target
def clean(fn):
ligands = os.path.join(os.path.dirname(fn), "..", "PDB",
"ligands_allgood.sdf")
if not os.path.isfile(ligands):
return
inchikeys = set(
[mol.write('inchikey') for mol in pybel.readfile('sdf', ligands)])
ofn = fn.replace('.sdf', '_filtered.sdf')
if not os.path.isfile(ofn):
good = 0
total = 0
o = rdk.Outputfile('sdf', ofn, overwrite=True)
for mol in pybel.readfile('sdf', fn):
total += 1
if mol.write('inchikey') not in inchikeys:
try:
o.write(rdk.readstring('mol', mol.write('mol')))
good += 1
except Exception, e:
#print e
pass
o.close()
print "%s %% remaining mols (%s/%s)" % (
(good * 100. / total), good, total)
def parse_decoy_files(decoyfilelist):
"""
Parses files containing known decoys
"""
decoy_set = set()
for decoyfile in decoyfilelist:
decoyfile = str(decoyfile)
mols = pybel.readfile(get_fileformat(decoyfile), decoyfile)
for mol in mols:
try:
cmol = ComparableMol(mol)
cmol.calcdesc()
decoy_set.add(cmol)
except Exception, e:
print e
continue
def parse_query_files(filelist):
"""
Parses files containing active ligands
"""
query_dict = {}
for file in filelist:
file = str(file)
mols = pybel.readfile(get_fileformat(file), file)
for mol in mols:
try:
cmol = ComparableMol(mol)
cmol.calcdesc()
query_dict[cmol] = 0
except Exception, e:
print e
continue
def parse_db_files(filelist):
"""
Parses files where to look for decoys
"""
filecount = 0
if type(filelist) == list:
random.shuffle(filelist)
for dbfile in filelist:
mols = pybel.readfile(get_fileformat(dbfile), dbfile)
for mol in mols:
try:
cmol = ComparableMol(mol)
except Exception, e:
print e
cmol = False
if cmol:
yield cmol, filecount, dbfile
filecount += 1
def validate(fn, typ):
ofn = fn.replace('.sdf.gz', '_sanitized.sdf')
if not os.path.isfile(ofn):
good = 0
total = 0
o = rdk.Outputfile('sdf', ofn, overwrite=True)
n = 0
for mol in pybel.readfile('sdf', fn):
mol.title = typ + "_" + mol.title + "_" + str(n)
total += 1
n += 1
try:
o.write(rdk.readstring('mol', mol.write('mol')))
good += 1
except Exception, e:
print e
pass
print "%s %% good mols (%s/%s)" % ((good * 100. / total), good, total)
o.close()
def read_fragment(self,
name=None,
fmt=None,
handle=None,
zero_to_origin=True):
"""Read a molecular structure from a file. Guess at the file type from
extension if caller does not supply explicit fmt. If file handle is
provided, read data from it. Otherwise open file name for reading.
:param name: file to open
:type name : str
:param fmt: optional OpenBabel format code e.g. "xyz"
:type fmt : str
:param zero_to_origin: translate geometry to put atom 0 at origin
:type zero_to_origin : bool
:return: molecular fragment
:rtype : Fragment
"""
if not fmt:
try:
fmt = name.rsplit(".", 1)[-1]
except (IndexError, AttributeError):
msg = "No fmt given for {0} and unable to guess from file extension".format(
repr(name))
if handle is None:
molecule = pybel.readfile(fmt, name).next()
else:
data = handle.read()
molecule = pybel.readstring(fmt, data)
fragment = Fragment(molecule)
if zero_to_origin:
fragment.set_zero_to_origin()
return fragment
def get_coord_dict(format, inputmol):
molH = pybel.readfile(format, inputmol).next()
molH.OBMol.DeleteHydrogens()
return {atom.idx: atom.coords for atom in molH}
elif n == 2:
if idx in coord_map:
pass
else:
coord_map.update({idx: ind})
elif n == 1:
if idx in coord_map:
pass
else:
coord_map.update({idx: ind})
else:
pass
if len(coord_map) == len(original_mol2):
coord_conform = {}
for index1, index2 in coord_map.items():
coord_conform.update({index1: docked_pdbqt.get(index2)})
mol2 = pybel.readfile('mol2', origanal_mol2_mol).next()
mol2.OBMol.DeleteHydrogens()
for atom in mol2:
atom.OBAtom.SetVector(
coord_conform.get(atom.idx)[0],
coord_conform.get(atom.idx)[1],
coord_conform.get(atom.idx)[2])
mol2.write('mol2', output_mol2, overwrite=True)
else:
print 'Lost coordinates in mapping'
else:
print 'Not equal number of atoms in molecules'
print docked_pdbqt_mol
def get_coord_dict(format, inputmol):
molH = pybel.readfile(format, inputmol).next()
molH.OBMol.DeleteHydrogens()
return {atom.idx: atom.coords for atom in molH}
if n == 3:
coord_map.update({idx:ind})
elif n == 2:
if idx in coord_map:
pass
else:
coord_map.update({idx:ind})
elif n == 1:
if idx in coord_map:
pass
else:
coord_map.update({idx:ind})
else:
pass
if len(coord_map) == len(original_mol2):
coord_conform = {}
for index1, index2 in coord_map.items():
coord_conform.update({index1:docked_pdbqt.get(index2)})
mol2 = pybel.readfile('mol2', origanal_mol2_mol).next()
mol2.OBMol.DeleteHydrogens()
for atom in mol2:
atom.OBAtom.SetVector(coord_conform.get(atom.idx)[0], coord_conform.get(atom.idx)[1], coord_conform.get(atom.idx)[2])
mol2.write('mol2', output_mol2, overwrite=True)
else:
print 'Lost coordinates in mapping'
else:
print 'Not equal number of atoms in molecules'
print docked_pdbqt_mol
