def collect_1(self, checkboard = []):
def getChains(s):
ret = s.split('___')[:2]
assert(len(ret) == 2)
return ret
parser = PDBParser()
io = PDBIO()
for f in self.files:
if not checkboard:
break
if f not in checkboard:
continue
try:
os.mkdir(os.path.join( self.outpath, f))
except OSError as e:
if e.errno != errno.EEXIST:
raise
structure = parser.get_structure(f, os.path.join(self.inpath, f))
chain_A, chain_B = getChains(f)
io.set_structure(structure[0]['A'])
io.save(os.path.join(self.outpath,f,chain_A + '.pdb'))
io.set_structure(structure[0]['B'])
io.save(os.path.join(self.outpath,f,chain_B + '.pdb'))
#make this module can be reuse to other application
self.then_do(f)
#remove the finished file from checkboard
checkboard.remove(f)
def __init__(self, structure_file, file_type=None):
PDBIO.__init__(self)
dirname, filename_without_extension, file_type2 = ssbio.utils.split_folder_and_path(
structure_file)
self.structure_file = structure_file
# Unzip the file if it is zipped
if file_type == '.gz':
unzipped = ssbio.utils.gunzip_file(structure_file, outdir=dirname)
dirname, filename_without_extension, file_type2 = ssbio.utils.split_folder_and_path(
unzipped)
if not file_type:
file_type = file_type2
else:
if '.' not in file_type:
file_type = '.{}'.format(file_type)
if file_type in ['.pdb', '.ent', '.mmcif', '.cif', '.mmtf']:
# Load the structure
if file_type.lower() == '.pdb' or file_type.lower() == '.ent':
structure = pdbp.get_structure(id='ssbio_pdb',
file=structure_file)
if file_type.lower() == '.mmcif' or file_type.lower() == '.cif':
with warnings.catch_warnings():
warnings.simplefilter('ignore', PDBConstructionWarning)
structure = cifp.get_structure(structure_id='ssbio_cif',
filename=structure_file)
if file_type.lower() == '.mmtf':
with warnings.catch_warnings():
warnings.simplefilter('ignore', PDBConstructionWarning)
structure = mmtfp.get_structure(file_path=structure_file)
log.debug('{}: parsed 3D coordinates of structure'.format(
op.basename(structure_file)))
else:
raise ValueError('{}: unsupported file type'.format(file_type))
# If there are multiple models (NMR), use the first model as the representative structure
# if len(structure) > 1:
# self.first_model = structure[0]
# structure = structure[0]
# self.set_structure(structure)
# log.debug('{}: using first model'.format(structure_file))
# elif len(structure) == 0:
# log.error('{}: no models in structure!'.format(structure_file))
# else:
self.set_structure(structure)
self.first_model = structure[0]
def internal_water_coordinates(self, file, get_normalized=False):
io = PDBIO()
parser = PDBParser(QUIET=True)
code = file[0:4]
struct = parser.get_structure(code, file)
plane = list(struct[0][' '].get_residues())
up_plane = plane[1]['O'].get_coord()[2]
down_plane = plane[0]['N'].get_coord()[2]
if len(list(list(struct[0])[0].get_residues())) > 100:
chain = list(struct[0])[0]
else:
chain = list(struct[0])[1]
chain_id = chain.get_id()
waters = water_in_chain(file, chain_id)
internal_waters = [
water for water in waters if water['O'].get_coord()[2] < up_plane
and water['O'].get_coord()[2] > down_plane
]
internal_water_coords = [
water['O'].get_coord() for water in internal_waters
]
if get_normalized:
norm_coords = internal_water_coords / up_plane
return norm_coords
else:
return internal_water_coords
def __init__(self, structure_file, file_type=None):
PDBIO.__init__(self)
dirname, filename_without_extension, file_type2 = ssbio.utils.split_folder_and_path(
structure_file)
self.structure_file = structure_file
# Unzip the file if it is zipped
if file_type == '.gz':
unzipped = ssbio.utils.gunzip_file(structure_file, outdir=dirname)
dirname, filename_without_extension, file_type2 = ssbio.utils.split_folder_and_path(
unzipped)
if not file_type:
file_type = file_type2
else:
if '.' not in file_type:
file_type = '.{}'.format(file_type)
if file_type.lower() in ['.pdb', '.ent', '.mmcif', '.cif', '.mmtf']:
# Load the structure
if file_type.lower() == '.pdb' or file_type.lower() == '.ent':
structure = pdbp.get_structure(id='ssbio_pdb',
file=structure_file)
if file_type.lower() == '.mmcif' or file_type.lower() == '.cif':
with warnings.catch_warnings():
warnings.simplefilter('ignore', PDBConstructionWarning)
structure = cifp.get_structure(structure_id='ssbio_cif',
filename=structure_file)
if file_type.lower() == '.mmtf':
with warnings.catch_warnings():
warnings.simplefilter('ignore', PDBConstructionWarning)
structure = mmtfp.get_structure(file_path=structure_file)
log.debug('{}: parsed 3D coordinates of structure'.format(
op.basename(structure_file)))
else:
raise ValueError('{}: unsupported file type'.format(file_type))
self.set_structure(structure)
try:
self.first_model = structure[0]
except KeyError:
raise KeyError(
'{}: no models contained in structure! Please check structure file contents.'
.format(structure_file))
def has_correct_dum_in_pdb(self, file_path):
io = PDBIO()
parser = PDBParser(QUIET=True)
code = file_path[-8:-4]
struct = parser.get_structure(code, file_path)
try:
list(struct[0][' '].get_residues())
except KeyError:
print('Error in file:', file_path)
def collect_0(self):
def getChains(s):
ret = s.split('___')[:2]
assert(len(ret) == 2)
return ret
parser = PDBParser()
io = PDBIO()
for f in self.files:
try:
os.mkdir(os.path.join( self.outpath, f))
except OSError as e:
if e.errno != errno.EEXIST:
raise
structure = parser.get_structure(f, os.path.join(self.inpath, f))
chain_A, chain_B = getChains(f)
io.set_structure(structure[0]['A'])
io.save(os.path.join(self.outpath,f,chain_A + '.pdb'))
io.set_structure(structure[0]['B'])
io.save(os.path.join(self.outpath,f,chain_B + '.pdb'))
self.then_do(f)
def write_pdb(structure,
file_name,
selector=None,
preserve_atom_numbering=False):
""" Write a PDB file from a given structure with a given file_name. Optionally, write specific atoms with selector and preserve atom numbering. """
writer = PDBIO()
writer.set_structure(structure)
if selector is None:
writer.save(file_name, preserve_atom_numbering=preserve_atom_numbering)
else:
writer.save(file_name,
selector,
preserve_atom_numbering=preserve_atom_numbering)
def save_structure(self, output_pdb_path: str, mod_id: str = None):
"""
Saves structure on disk in PDB format
Args:
output_pdb_path: OS path to the output file
mod_id (optional): model to write
Errors:
OSError: Error saving the file
"""
if not output_pdb_path:
raise OutputPathNotProvidedError
pdbio = PDBIO()
if mod_id is None:
pdbio.set_structure(self.st)
pdbio.save(output_pdb_path)
else:
pdbio.set_structure(self.st[mod_id])
pdbio.save(output_pdb_path)
def CreatePDB(self, coordArray, fPath, ofile):
sloppyparser = PDBParser(PERMISSIVE=True, QUIET=True)
structure = sloppyparser.get_structure("MD_system", fPath)
print("\nGenerating PDB file...")
sb = StructureBuilder()
sb.set_header(structure.header)
# Iterate through models
for i in range(len(list(structure.get_models()))):
# Iterate through chains
models = list(structure.get_models())
counter = 0
for j in range(len(list(models[i].get_chains()))):
chains = list(models[i].get_chains())
#Iterate thgouth residues
for k in range(len(list(chains[j].get_residues()))):
#Iterate through
residues = list(chains[j].get_residues())
for l in range(len(list(residues[k].get_atoms()))):
#Set coord for each
for atom in structure[i][chains[j].id][
residues[k].id].get_atoms():
structure[i][chains[j].id][residues[k].id][
atom.id].set_coord(
np.array((float(coordArray[counter][0]),
float(coordArray[counter][1]),
float(coordArray[counter][2]))))
#print(structure[i][chains[j].id][residues[k].id][atom.id].get_vector())
counter += 1
io = PDBIO()
io.set_structure(structure)
io.save(ofile)
print("Transform file written to: " + ofile)
def write_PDB(entity: Structure,
file: str,
pdbid: str = None,
chainid: str = None) -> None:
"""Write PDB file with HEADER and TITLE."""
enumerate_atoms(entity)
with as_handle(file, "w") as fp:
try:
if "S" == entity.level:
if hasattr(entity, "header"):
if not pdbid:
pdbid = entity.header.get("idcode", None)
hdr = entity.header.get("head", None)
dd = pdb_date(entity.header.get("deposition_date", None))
if hdr:
fp.write(("HEADER {:40}{:8} {:4}\n").format(
hdr.upper(), (dd or ""), (pdbid or "")))
nam = entity.header.get("name", None)
if nam:
fp.write("TITLE " + nam.upper() + "\n")
io = PDBIO()
io.set_structure(entity)
io.save(fp, preserve_atom_numbering=True)
else:
raise PDBException("level not 'S': " + str(entity.level))
except KeyError:
raise Exception(
"write_PIC: argument is not a Biopython PDB Entity " +
str(entity))
def cut_7_helix(self, source_pdb, target_folder, offset=0):
io = PDBIO()
parser = PDBParser(QUIET=True)
code = source_pdb[-8:-4]
struct = parser.get_structure(code, source_pdb)
plane = list(struct[0][' '].get_residues())
try:
up_plane = plane[1]['O'].get_coord()[2] + offset
except KeyError:
up_plane = plane[0]['O'].get_coord()[2] + offset
try:
down_plane = plane[0]['N'].get_coord()[2] - offset
except KeyError:
down_plane = plane[1]['N'].get_coord()[2] - offset
chain_id = self.get_TM_chain(list(struct[0]), up_plane, down_plane)
chain = struct[0][chain_id]
io.set_structure(struct)
io.save(target_folder + code + '_tm.pdb',
self.TMSelect(up_plane, down_plane, chain))
def scwrl(self, altseq):
""" Repacks sidechains using SCWRL4 and returns a copy """
io = PDBIO()
seqfname = "temp/%d.txt" % multidigit_rand(10)
with open(seqfname, 'wb') as seqfile:
structfile = "temp/%d.pdb" % multidigit_rand(10)
seqfile.write(altseq)
scwrlfile = structfile + ".scwrl"
io.set_structure(self.structure)
io.save(structfile)
cmd = [
"scwrl", "-0", "-i", structfile, '-s', seqfname, '-o', scwrlfile
]
print "\n%s" % ' '.join(cmd)
sp.Popen(cmd, stdout=sp.PIPE, stderr=sp.PIPE).communicate()
p = PDBParser()
with open(scwrlfile, 'rb') as fin:
filterwarnings('ignore', category=PDBConstructionWarning)
s = p.get_structure(self.id, scwrlfile)
resetwarnings()
s = PDBMapStructure(s, pdb2pose={}, refseq=self.refseq)
os.remove(structfile)
os.remove(scwrlfile)
os.remove(seqfname)
return s
def test_pdbio_write_pqr_structure(self):
"""Write a full structure using PDBIO."""
# Create a PDBIO object in pqr mode with example_structure as an argument
io = PDBIO(is_pqr=True)
io.set_structure(self.example_structure)
# Write to a temporary file
filenumber, filename = tempfile.mkstemp()
os.close(filenumber)
try:
# Export example_structure to a temp file
io.save(filename)
# Parse exported structure
output_struct = self.pqr_parser.get_structure("1a8o", filename)
# Comparisons
self.assertEqual(len(output_struct),
len(self.example_structure)) # Structure Length
original_residues = len(list(
self.example_structure.get_residues()))
parsed_residues = len(list(output_struct.get_residues()))
self.assertEqual(parsed_residues,
original_residues) # Number of Residues
# Atom-wise comparison
original_atoms = self.example_structure.get_atoms()
for atom in output_struct.get_atoms():
self.assertEqual(atom, next(original_atoms))
finally:
os.remove(filename)
def main(pdbfile, scheme, outfile):
pdb_io = PDBIO()
parser = PDBParser()
structure = parser.get_structure('self', pdbfile)
model = structure[0]
chain = ' '
anarci_dict = {}
fastafile = Path(pdbfile).stem + '.fasta'
with open(fastafile, 'w+') as ff:
subprocess.run(['pdb_tofasta', '-multi', pdbfile], stdout=ff)
out = subprocess.run(['anarci', '-i', fastafile, '--scheme', scheme],
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT)
output = out.stdout.decode("utf-8").splitlines()
for line in output:
l = line.strip()
li = line.split()
if line.startswith('#'):
if re.search("PDB", l):
ch = l.split('|')
chain = ch[1]
anarci_dict[chain] = []
elif line.startswith(chain):
if len(li) == 3:
if li[2] == '-':
continue
else:
ch = li[0]
resseq = int(li[1])
inscode = ' '
residue = li[2]
het = ' '
tuple = (het, resseq, inscode)
anarci_dict[chain].append(tuple)
elif len(li) == 4:
ch = li[0]
resseq = int(li[1])
inscode = str(li[2])
residue = li[3]
het = ' '
tuple = (het, resseq, inscode)
anarci_dict[chain].append(tuple)
else:
continue
for residue in structure.get_residues():
residue.id = (' ', residue.id[1] + 900, residue.id[2])
for d in anarci_dict:
count = 0
for i, residue in enumerate(model[d]):
if i < len(anarci_dict[d]):
residue.id = anarci_dict[d][i]
count = residue.id[1]
else:
count = count + 1
residue.id = (' ', count, ' ')
pdb_io.set_structure(structure)
pdb_io.save(outfile)
def write_PDB(entity, file, pdbid=None, chainid=None):
"""Write PDB file with HEADER and TITLE."""
with as_handle(file, 'w') as fp:
try:
if 'S' == entity.level:
if not pdbid:
pdbid = entity.header.get('idcode', None)
hdr = entity.header.get('head', None)
dd = entity.header.get('deposition_date', None)
if hdr:
fp.write(('HEADER {:40}{:8} {:4}\n'
).format(hdr.upper(), (dd or ''), (pdbid or '')))
nam = entity.header.get('name', None)
if nam:
fp.write('TITLE ' + nam.upper() + '\n')
io = PDBIO()
io.set_structure(entity)
io.save(fp)
else:
raise PDBException("level not 'S': "
+ str(entity.level))
except KeyError:
raise Exception(
"write_PIC: argument is not a Biopython PDB Entity "
+ str(entity))
def run_naccess(model,
pdb_file,
probe_size=None,
z_slice=None,
naccess='naccess',
temp_path='/tmp/'):
# make temp directory;
tmp_path = tempfile.mkdtemp(dir=temp_path)
# file name must end with '.pdb' to work with NACCESS
# -> create temp file of existing pdb
# or write model to temp file
handle, tmp_pdb_file = tempfile.mkstemp('.pdb', dir=tmp_path)
os.close(handle)
if pdb_file:
pdb_file = os.path.abspath(pdb_file)
shutil.copy(pdb_file, tmp_pdb_file)
else:
writer = PDBIO()
writer.set_structure(model.get_parent())
writer.save(tmp_pdb_file)
# chdir to temp directory, as NACCESS writes to current working directory
old_dir = os.getcwd()
os.chdir(tmp_path)
# create the command line and run
# catch standard out & err
command = [naccess, tmp_pdb_file]
if probe_size:
command.extend(['-p', probe_size])
if z_slice:
command.extend(['-z', z_slice])
p = subprocess.Popen(command,
universal_newlines=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
out, err = p.communicate()
os.chdir(old_dir)
rsa_file = tmp_pdb_file[:-4] + '.rsa'
asa_file = tmp_pdb_file[:-4] + '.asa'
# Alert user for errors
if err.strip():
warnings.warn(err)
if (not os.path.exists(rsa_file)) or (not os.path.exists(asa_file)):
raise Exception('NACCESS did not execute or finish properly.')
# get the output, then delete the temp directory
with open(rsa_file) as rf:
rsa_data = rf.readlines()
with open(asa_file) as af:
asa_data = af.readlines()
# shutil.rmtree(tmp_path, ignore_errors=True)
return rsa_data, asa_data
def writeFileBioPDB(struct,fn): io = PDBIO() io.set_structure(struct) try: io.save(fn) except Exception, e: print " "+str(e) sys.exit(1)
def save_chain_to(chain, filename: str):
from Bio.PDB.PDBIO import PDBIO
io = PDBIO()
# io.set_structure(chain.get_bio_chain())
structure = Structure(filename)
structure.add(chain)
io.set_structure(structure)
io.save(filename)
def extract(structure, chain_id, start, end, filename):
"""
Write out selected portion to filename.
"""
sel = ChainSelector(chain_id, start, end)
io = PDBIO()
io.set_structure(structure)
io.save(filename, sel)
def filter_pdb(input_path, output_path, chain):
"""
Filter a PDB file to the chain of interest
"""
pdb_name = Path(input_path).stem
pdb_parser = PDBParser()
structure = pdb_parser.get_structure(pdb_name, input_path)
pdbio = PDBIO()
pdbio.set_structure(structure)
pdbio.save(output_path, select=ChainSelect(chain))
def export_structure(structure, name, format):
"""
Writes the strucuture into a file. The file can be either a pdb or a mmcif.
"""
if format == "pdb":
io = PDBIO()
elif format == "cif":
io = pdb.MMCIFIO()
io.set_structure(structure)
io.save(name)
def remove_residues(self, request, claims):
"""
Remove residues from a PDB structure
For a detailed input description see the file:
mdstudio_structures/schemas/endpoints/removed_residues_request_v1.json
And for a detailed description of the output see:
mdstudio_structures/schemas/endpoints/removed_residues_response_v1.json
"""
request['workdir'] = os.path.abspath(request['workdir'])
# Parse the structure
parser = PDBParser(PERMISSIVE=True)
struc_obj = StringIO(request.get('mol'))
structure = parser.get_structure('mol_object', struc_obj)
struc_obj.close()
to_remove = [r.upper() for r in request.get('residues', [])]
removed = []
for model in structure:
for chain in model:
for residue in chain:
if residue.get_resname() in to_remove:
chain.detach_child(residue.id)
removed.append(residue.get_resname())
if len(chain) == 0:
model.detach_child(chain.id)
self.log.info('Removed residues: {0}'.format(','.join(removed)))
# Save to file or string
pdbio = PDBIO()
pdbio.set_structure(structure)
status = 'completed'
if request.get('workdir'):
result = os.path.join(request.get('workdir'), 'structure.pdb')
pdbio.save(result)
else:
outfile = StringIO()
pdbio.save(outfile)
outfile.seek(0)
result = outfile.read()
return {'status': status, 'mol': result}
def pose(self):
""" Loads the PDBMapStructure as a Rosetta::Pose object """
import_rosetta()
io = PDBIO()
io.set_structure(self.structure)
with tempfile.NamedTemporaryFile('wrb',suffix='.pdb',delete=False) as tf:
io.save(tf.name)
pose = rosetta.Pose()
rosetta.pose_from_pdb(pose,tf.name)
os.remove(tf.name)
return pose
def get_pdb_string(pdb_path, select=None):
"""
Get string representation of a PDB file, filtered using select as in Bio.PDB.PDBIO
"""
pdb_parser = PDBParser()
structure = pdb_parser.get_structure('_', pdb_path)
pdbio = PDBIO()
pdbio.set_structure(structure)
with StringIO('PDB') as virtual_pdb_file:
pdbio.save(virtual_pdb_file, select=select)
return virtual_pdb_file.getvalue()
def run_naccess(model,
pdb_file,
probe_size=None,
z_slice=None,
naccess='naccess',
temp_path='/tmp/'):
# make temp directory; chdir to temp directory,
# as NACCESS writes to current working directory
tmp_path = tempfile.mktemp(dir=temp_path)
os.mkdir(tmp_path)
old_dir = os.getcwd()
os.chdir(tmp_path)
# file name must end with '.pdb' to work with NACCESS
# -> create temp file of existing pdb
# or write model to temp file
tmp_pdb_file = tempfile.mktemp('.pdb', dir=tmp_path)
if pdb_file:
os.system('cp %s %s' % (pdb_file, tmp_pdb_file))
else:
writer = PDBIO()
writer.set_structure(model.get_parent())
writer.save(tmp_pdb_file)
# create the command line and run
# catch standard out & err
command = '%s %s ' % (naccess, tmp_pdb_file)
if probe_size:
command += '-p %s ' % probe_size
if z_slice:
command += '-z %s ' % z_slice
in_, out, err = os.popen3(command)
in_.close()
stdout = out.readlines()
out.close()
stderr = err.readlines()
err.close()
# get the output, then delete the temp directory
rsa_file = tmp_pdb_file[:-4] + '.rsa'
rf = open(rsa_file)
rsa_data = rf.readlines()
rf.close()
asa_file = tmp_pdb_file[:-4] + '.asa'
af = open(asa_file)
asa_data = af.readlines()
af.close()
os.chdir(old_dir)
os.system('rm -rf %s >& /dev/null' % tmp_path)
return rsa_data, asa_data
def main():
if len(sys.argv) < 2:
print("Please input fasta filename as an argument for the script")
sys.exit()
filename = sys.argv[1]
if filename.endswith('.fasta'):
blast_result = blast_sequence(filename)
else:
print("Unknown file format (need to be fasta file)")
sys.exit()
pdb_filepath = download_pdb_support_sequence(blast_result)
filter_pdb(pdb_filepath)
p = PDBParser(PERMISSIVE=1)
structure = p.get_structure('file', 'filtered.pdb')
seqObj = generate_fasta_from_pdb(structure)
target_seq = SeqIO.read(filename, format="fasta").seq
alignments = pairwise_alignement(target_seq, seqObj.seq)
#Consider CA atoms only for computing distances
ca_atoms = [
atom for model in structure for chain in model for residue in chain
for atom in residue if atom.get_id() == "CA"
]
gaps = insert_gaps_in_atomlist(ca_atoms, alignments[0][0])
insertions = insert_gaps_in_atomlist(ca_atoms, alignments[0][1])
print(gaps)
print(insertions)
print("Searching for best part for gaps")
gap_parts = get_gaps_parts(gaps)
print("Searching for best part for insertions")
ins_parts = get_gaps_parts(
insertions, get_insertion_type(alignments[0][0], alignments[0][1]))
print("Completing gaps/insertions by found parts")
complete_atoms = [
atom for model in structure for chain in model for residue in chain
for atom in residue
]
complete_atoms = insert_gaps_insertions_in_atomlist(
complete_atoms, ins_parts, gap_parts)
print("Writing final pdb file in : final.pdb")
io = PDBIO()
io.set_structure(build_structure(complete_atoms))
io.save('final.pdb')
def generate_output_file(final_model, out_name):
""" This function takes as input both the final model created with the building algorithm and the output filename given by the user (if not defined, is macrocomplex by default). Eventually, it returns the file saved in either ".pdb" or ".mmcif" format. """
out_name = str(out_name.strip())
# If the output file is too big, we save it in ".mmcif" format
if len(list(final_model[0].get_atoms())) > 99999 or len(
list(final_model[0].get_chains())) > 62:
mmcif_IO = MMCIFIO()
mmcif_IO.set_structure(final_model[0])
mmcif_IO.save(out_name + ".cif")
# Otherwise, save it ".pdb" format
else:
pdb_IO = PDBIO()
pdb_IO.set_structure(final_model[0])
pdb_IO.save(out_name + ".pdb")
def clean(self):
p = PDBParser()
io = PDBIO()
with tempfile.NamedTemporaryFile('wrb',suffix='.pdb',delete=False) as tf:
io.set_structure(self.structure)
io.save(tf.name)
cmd = ['lib/clean_pdb.py',tf.name,'ignorechain','nopdbout']
logger.info("Shell to: %s"%' '.join(cmd))
proc = sp.Popen(cmd,stdout=sp.PIPE)
s = p.get_structure(self.get_id(),proc.stdout)
p = lib.PDBMapIO.PDBMapParser()
s = p.process_structure(s,force=True)
s = PDBMapStructure(s,pdb2pose=self._pdb2pose,refseq=self.refseq)
os.remove(tf.name)
return s
def main(args):
"""Main script"""
pdb_name = Path(args.pdb).stem
# deal with FoldX repaired PDBs
if pdb_name.endswith('_Repair'):
pdb_name = pdb_name.replace('_Repair', '')
pdb_parser = PDBParser()
structure = pdb_parser.get_structure(pdb_name, args.pdb)
sections = import_sections(args.yaml, pdb_name)
pdbio = PDBIO()
pdbio.set_structure(structure)
pdbio.save(sys.stdout, select=SectionSelecter(sections))
def func1():
import sys
import re
import gzip
from Bio.PDB.MMCIFParser import MMCIFParser
parser = MMCIFParser(QUIET=True)
from Bio.PDB.PDBParser import PDBParser
parser1 = PDBParser(PERMISSIVE=0, QUIET=True)
from Bio.PDB.PDBIO import PDBIO
#pathmmcif = "/Users/tarun/Documents/mmCIF"
#pathmmcif = "/data/pdb/divided/mmCIF"
pathmmcif = "/Volumes/BIOINFO/mmCIF"
#pathmmcif = "/Volumes/RCSB_DATA/pdb"
#count = 0
#if count == 0:
try:
pdb1 = "{}".format(sys.argv[2])
fol = pdb1[1:3]
c1 = "{}".format(sys.argv[3])
pdbfile = "{}/{}/{}.cif.gz".format(pathmmcif, fol, pdb1)
#pdbfile = "{}/{}/pdb{}.ent.gz".format(pathmmcif,fol,pdb1)
tar = gzip.open("{}".format(pdbfile), "rb")
out = open("pdbprocess.cif", "wb")
#out = open("pdbprocess.pdb","wb")
out.write(tar.read())
tar.close()
out.close()
structure_id = "{}".format(pdb1)
filename = "pdbprocess.cif"
#filename = "pdbprocess.pdb"
structure = parser.get_structure(structure_id, filename)
model = structure[0]
chain = model["{}".format(c1)]
io = PDBIO()
io.set_structure(chain)
io.save("chain1.pdb")
except:
print("FILE NOT FOUND")