def parse_options():
'''
'''
# PARSE COMMAND LINE
usage = "%prog [options]"
parser = OptionParser(usage=usage)
parser.add_option(
"--debug",
action="store_true",
dest="debug",
default=False,
help='Set logging level to debug and print more verbose output.')
parser.add_option(
"-B",
"--binsize",
dest="binsize",
type=float,
default=0.0,
help=
"Bin size (in Angstrom) to use for binning contacts based on inter-atomic distance."
)
parser.add_option(
"-F",
"--format",
dest="format",
default='csv',
help="Format to use for writing the SIFt of the protein-ligand complex."
)
parser.add_option(
"-O",
"--output",
dest="output",
default=None,
help="File to which the data will be written (default=STDOUT).")
parser.add_option(
"-D",
"--descriptor",
dest="descriptor",
default='elements',
help=
"Descriptor to use. Valid descriptors are 'credo', 'elements' and 'sybyl'."
)
# GET COMMAND LINE OPTIONS
(options, args) = parser.parse_args()
if options.descriptor not in ('elements', 'credo', 'sybyl'):
logger.fatal("Invalid descriptor: {0}.".format(options.descriptor))
parser.print_help()
sys.exit(1)
return options
def parse_options():
'''
'''
# PARSE COMMAND LINE
usage = "%prog [options]"
parser = OptionParser(usage=usage)
parser.add_option("--debug",
action = "store_true",
dest = "debug",
default = False,
help = 'Set logging level to debug and print more verbose output.')
parser.add_option("-B", "--binsize",
dest = "binsize",
type = float,
default = 0.0,
help = "Bin size (in Angstrom) to use for binning contacts based on inter-atomic distance.")
parser.add_option("-F", "--format",
dest = "format",
default = 'csv',
help = "Format to use for writing the SIFt of the protein-ligand complex.")
parser.add_option("-O", "--output",
dest = "output",
default = None,
help = "File to which the data will be written (default=STDOUT).")
parser.add_option("-D", "--descriptor",
dest = "descriptor",
default = 'elements',
help = "Descriptor to use. Valid descriptors are 'credo', 'elements' and 'sybyl'.")
# GET COMMAND LINE OPTIONS
(options, args) = parser.parse_args()
if options.descriptor not in ('elements', 'credo', 'sybyl'):
logger.fatal("Invalid descriptor: {0}.".format(options.descriptor))
parser.print_help()
sys.exit(1)
return options
def main():
'''
'''
options = parse_options()
# THIS OPTION WILL PRODUCE MORE VERBOSE OUTPUT
if options.debug: logger.setLevel(logging.DEBUG)
if options.output: fh = open(options.output,'wb')
else: fh = sys.stdout
# CHOOSE HOW THE OUPTPUT DATA WILL BE WRITTEN
if options.format == 'csv':
writer = csv.writer(fh, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
HEADER = True
# iterate through all mol2 files in inputdir
for mol2file in glob.glob(os.path.join(options.inputligdir, '*.mol2')):
#lig_path = os.path.join(option.inputdir, file)
lig_path = mol2file
protein_path = options.inputpdb
if not os.path.isfile(protein_path):
logger.fatal("The protein file does not exist.".format(options.inputpdb))
sys.exit(1)
protein = get_molecule(protein_path)
ligand = get_molecule(lig_path)
# calculate descriptor based on the sum of interacting element pairs
if options.descriptor == 'elements':
# calculate element pair descriptor for this complex
descriptor, labels = contacts.element_descriptor(protein, ligand,
binsize=options.binsize)
# calculate descriptor based on the sum of interacting element pairs
elif options.descriptor == 'sybyl':
# calculate element pair descriptor for this complex
descriptor, labels = contacts.sybyl_atom_type_descriptor(protein, ligand,
binsize=options.binsize)
# calculate descriptor using structural interaction fingerprints
elif options.descriptor == 'credo':
# get the protein-ligand structural interaction fingerprint
descriptor, labels = contacts.sift_descriptor(protein, ligand,
binsize=options.binsize)
if HEADER:
# UPDATE COLUMN LABELS
labels.append('ligand')
writer.writerow(labels)
HEADER = False
if options.format == 'csv':
#ligandname = "\"" + os.path.basename(lig_path).split('.')[0] + "\""
ligandname = os.path.basename(lig_path).split('.')[0]
#print(ligandname)
# FIRST COLUMN OF OUTPUT ROW
row = descriptor.tolist() + [ligandname]
writer.writerow(row)
def main():
"""
"""
options = parse_options()
# this option will produce more verbose output
if options.debug: logger.setLevel(logging.DEBUG)
csarconf = config['csar']
if options.output: fh = open(options.output,'wb')
else: fh = sys.stdout
# choose how the ouptput data will be written
if options.format == 'csv':
writer = csv.writer(fh, delimiter=',', quotechar='"',
quoting=csv.QUOTE_MINIMAL)
HEADER = True
# iterate through all numbered directories
for directory in os.listdir(csarconf['directory']):
entrydir = os.path.join(csarconf['directory'], directory)
# parse kd.dat to get the pKd
kddat_path = os.path.join(entrydir, 'kd.dat')
# exit if kd.dat is missing
if not os.path.isfile(kddat_path):
logger.fatal("CSAR directory {} does not contain kd.dat file."
.format(directory))
sys.exit(1)
entry, pdb, pkd = open(kddat_path).read().strip().replace(' ','').split(',')
protein_path = glob.glob(os.path.join(entrydir, '*_complex.mol2')).pop()
protein = get_molecule(str(protein_path))
ligand = extract_ligand(protein.OBMol)
# calculate descriptor based on the sum of interacting element pairs
if options.descriptor == 'elements':
# calculate element pair descriptor for this complex
descriptor, labels = contacts.element_descriptor(protein, ligand,
binsize=options.binsize)
# calculate descriptor based on the sum of interacting element pairs
elif options.descriptor == 'sybyl':
# calculate element pair descriptor for this complex
descriptor, labels = contacts.sybyl_atom_type_descriptor(protein, ligand,
binsize=options.binsize)
# calculate descriptor using structural interaction fingerprints
elif options.descriptor == 'credo':
# get the protein-ligand structural interaction fingerprint
descriptor, labels = contacts.sift_descriptor(protein, ligand,
binsize=options.binsize)
if HEADER:
# UPDATE COLUMN LABELS
labels.insert(0,'pKd/pKi')
labels.append('pdb')
writer.writerow(labels)
HEADER = False
if options.format == 'csv':
# FIRST COLUMN OF OUTPUT ROW
row = [pkd] + descriptor.tolist() + [pdb]
writer.writerow(row)
def parse_options():
'''
'''
# PARSE COMMAND LINE
usage = "%prog [options]"
parser = OptionParser(usage=usage)
parser.add_option("--debug",
action = "store_true",
dest = "debug",
default = False,
help = 'Set logging level to debug and print more verbose output.')
parser.add_option("-B", "--binsize",
dest = "binsize",
type = float,
default = 0.0,
help = "Bin size (in Angstrom) to use for binning contacts based on inter-atomic distance.")
parser.add_option("-F", "--format",
dest = "format",
default = 'csv',
help = "Format to use for writing the SIFt of the protein-ligand complex.")
parser.add_option("-O", "--output",
dest = "output",
default = None,
help = "File to which the data will be written (default=STDOUT).")
parser.add_option("-P", "--pdbbind-dir",
dest = "pdbbind",
default = None,
help = "PDBbind directory.")
parser.add_option("-I", "--index",
dest = "index",
default = None,
help = "PDBbind data index file for a specific data set (core,refined,general).")
parser.add_option("-D", "--descriptor",
dest = "descriptor",
default = 'credo',
help = "Descriptor to use. Valid descriptors are 'credo', 'elements' and 'sybyl'.")
# GET COMMAND LINE OPTIONS
(options, args) = parser.parse_args()
if not options.pdbbind:
logger.error("The PDBbind directory must be provided.")
parser.print_help()
sys.exit(1)
elif not os.path.exists(options.pdbbind):
logger.fatal("The specified PDBbind directory does not exist.")
sys.exit(1)
if not options.index:
logger.error("A path to a PDBbind data index file must be provided.")
parser.print_help()
sys.exit(1)
elif not os.path.exists(options.index):
logger.fatal("The specified PDBbind data index file does not exist.")
sys.exit(1)
if options.descriptor not in ('elements', 'credo', 'sybyl'):
logger.fatal("Invalid descriptor: {0}.".format(options.descriptor))
parser.print_help()
sys.exit(1)
return options
def main():
"""
"""
options = parse_options()
# this option will produce more verbose output
if options.debug: logger.setLevel(logging.DEBUG)
csarconf = config['csar']
if options.output: fh = open(options.output, 'wb')
else: fh = sys.stdout
# choose how the ouptput data will be written
if options.format == 'csv':
writer = csv.writer(fh,
delimiter=',',
quotechar='"',
quoting=csv.QUOTE_MINIMAL)
HEADER = True
# iterate through all numbered directories
for directory in os.listdir(csarconf['directory']):
entrydir = os.path.join(csarconf['directory'], directory)
# parse kd.dat to get the pKd
kddat_path = os.path.join(entrydir, 'kd.dat')
# exit if kd.dat is missing
if not os.path.isfile(kddat_path):
logger.fatal(
"CSAR directory {} does not contain kd.dat file.".format(
directory))
sys.exit(1)
entry, pdb, pkd = open(kddat_path).read().strip().replace(
' ', '').split(',')
protein_path = glob.glob(os.path.join(entrydir,
'*_complex.mol2')).pop()
protein = get_molecule(str(protein_path))
ligand = extract_ligand(protein.OBMol)
# calculate descriptor based on the sum of interacting element pairs
if options.descriptor == 'elements':
# calculate element pair descriptor for this complex
descriptor, labels = contacts.element_descriptor(
protein, ligand, binsize=options.binsize)
# calculate descriptor based on the sum of interacting element pairs
elif options.descriptor == 'sybyl':
# calculate element pair descriptor for this complex
descriptor, labels = contacts.sybyl_atom_type_descriptor(
protein, ligand, binsize=options.binsize)
# calculate descriptor using structural interaction fingerprints
elif options.descriptor == 'credo':
# get the protein-ligand structural interaction fingerprint
descriptor, labels = contacts.sift_descriptor(
protein, ligand, binsize=options.binsize)
if HEADER:
# UPDATE COLUMN LABELS
labels.insert(0, 'pKd/pKi')
labels.append('pdb')
writer.writerow(labels)
HEADER = False
if options.format == 'csv':
# FIRST COLUMN OF OUTPUT ROW
row = [pkd] + descriptor.tolist() + [pdb]
writer.writerow(row)
def parse_options():
'''
'''
# PARSE COMMAND LINE
usage = "%prog [options]"
parser = OptionParser(usage=usage)
parser.add_option("--debug",
action = "store_true",
dest = "debug",
default = False,
help = 'Set logging level to debug and print more verbose output.')
parser.add_option("-B", "--binsize",
dest = "binsize",
type = float,
default = 0.0,
help = "Bin size (in Angstrom) to use for binning contacts based on inter-atomic distance.")
parser.add_option("-F", "--format",
dest = "format",
default = 'csv',
help = "Format to use for writing the SIFt of the protein-ligand complex.")
parser.add_option("-O", "--output",
dest = "output",
default = "/home/dat/WORK/DB/DESCRIPTORS/CASF2014-refined_SIFt_RMSD.csv",#None,
help = "File to which the data will be written (default=STDOUT).")
parser.add_option("-P", "--pdbbind-dir",
dest = "pdbbind",
default = "/home/dat/WORK/DB/PDBbind/v2014-refined/",#None,
help = "PDBbind directory.")
parser.add_option("-I", "--index",
dest = "index",
default = "/home/dat/WORK/DB/PDBbind/v2014-refined/INDEX_refined_data.2014",#None,
help = "PDBbind data index file for a specific data set (core,refined,general).")
parser.add_option("-D", "--descriptor",
dest = "descriptor",
default = 'credo',
help = "Descriptor to use. Valid descriptors are 'credo', 'elements' and 'sybyl'.")
# GET COMMAND LINE OPTIONS
(options, args) = parser.parse_args()
if not options.pdbbind:
logger.error("The PDBbind directory must be provided.")
parser.print_help()
sys.exit(1)
elif not os.path.exists(options.pdbbind):
logger.fatal("The specified PDBbind directory does not exist.")
sys.exit(1)
if not options.index:
logger.error("A path to a PDBbind data index file must be provided.")
parser.print_help()
sys.exit(1)
elif not os.path.exists(options.index):
logger.fatal("The specified PDBbind data index file does not exist.")
sys.exit(1)
if options.descriptor not in ('elements', 'credo', 'sybyl'):
logger.fatal("Invalid descriptor: {0}.".format(options.descriptor))
parser.print_help()
sys.exit(1)
return options