def write(infile, ftype, indata, ck=False):
"""
:param infile: Path to input file.
:type infile: str
:param ftype: One of 'psicov', 'ccmpred', 'fasta', 'pdb', 'a3m', 'jones', 'xml'.
:type ftype: str
:param ck: Open alternative conkit version instead of default, defaults to False.
:type ck: bool, optional
:return: Parsed file (and, for 'pdb', list of filenames).
:rtype: One or two of list[str], :class:`~crops.elements.sequences.sequence`, :class:`~conkit.core.sequence.Sequence`,
"""
if (ftype.lower() not in _ftypelist() or
isinstance(ftype, str) is not True):
logging.critical('Specified type not valid.')
raise ValueError
if ck is True and ftype.lower() != 'xml':
output = ckio.write(infile, ftype.lower(), hyerarchy=indata)
else:
if ftype.lower() == 'psicov':
pass
if ftype.lower() == 'ccmpred':
pass
elif ftype.lower() == 'fasta':
output = cps.parseseqfile(infile)
elif ftype.lower() == 'pdb':
output1, output2 = cps.parsestrfile(infile)
return output1, output2
elif ftype.lower() == 'a3m' or 'jones':
output = ckio.read(infile, ftype.lower())
elif ftype.lower() == 'xml':
output = ET.parse(infile)
return output
def main():
"""Remove a number of residues from sequence and structure files in agreement to the intervals and other details supplied.
:raises ValueError: For wrong argument values.
"""
# INITIALISE AND PARSE ARGUMENTS FROM COMMAND LINE
parser = create_argument_parser()
args = parser.parse_args()
global logger
logger = ccl.crops_logger(level="info")
logger.info(ccl._welcome())
inseq = check_path(args.input_seqpath[0], 'file')
indb = check_path(args.input_database[0], 'file')
instr = check_path(args.input_strpath[0])
if args.uniprot_threshold is not None:
if args.uniprot_threshold[1] != 'server-only':
insprot = check_path(args.uniprot_threshold[1])
else:
insprot = 'server-only'
else:
insprot = None
if args.uniprot_threshold is not None:
minlen = float(args.uniprot_threshold[0])
if minlen < 0.0 or minlen > 100.0:
logger.critical(
'The UniProt threshold is a percentage and, therefore, it must fulfil 0 < threshold < 100.'
)
raise ValueError
else:
minlen = 0.0
targetlbl = ctg.target_format(indb, terms=args.terminals, th=minlen)
infixlbl = ctg.infix_gen(indb, terms=args.terminals)
if args.outdir is None:
outdir = check_path(os.path.dirname(inseq), 'dir')
else:
outdir = check_path(os.path.join(args.outdir[0], ''), 'dir')
# PARSE INPUT FILES
logger.info('Parsing sequence file ' + inseq)
if args.preselect is not None:
subset = set(args.preselect)
else:
subset = None
seqset = cin.parseseqfile(seq_input=inseq, inset=subset)
logger.info('Done')
logger.info('Parsing structure file ' + instr)
strset, fileset = cin.parsestrfile(instr)
logger.info('Done')
logger.info('Parsing interval database file ' + indb)
if len(seqset) > 0:
intervals = cin.import_db(indb, pdb_in=seqset)
else:
logger.critical('No chains were imported from sequence file.')
raise ValueError
logger.info('Done' + os.linesep)
if insprot is not None and minlen > 0.0:
logger.info('Parsing uniprot sequence file: ' + insprot)
uniprotset = set()
for seqncid, seqnc in seqset.items():
chains = seqnc.chainlist()
for monomerid in chains:
monomer = seqnc.imer[seqnc.whatseq(monomerid)]
if 'uniprot' in intervals[seqncid][monomerid].tags:
for key in intervals[seqncid][monomerid].tags['uniprot']:
if key.upper() not in uniprotset:
uniprotset.add(key.upper())
upserver = True if insprot == 'server-only' else False
uniprotset = cin.parseseqfile(seq_input=insprot,
inset=uniprotset,
use_UPserver=upserver)
logger.info('Done' + os.linesep)
# MAIN OPERATION / PRINT OUT RESULTS WITHIN
gseqset = {}
strset2 = {}
logger.info('Renumbering structure(s)...')
for key, structure in strset.items():
found = False
for seqname in seqset:
if ((seqname in key) or (len(seqset) == 1 and len(strset) == 1)):
finalid = seqname
newstructure, gseqset[seqname] = cop.renumber_pdb(
seqset[seqname], structure, seqback=True)
fout = finalid + infixlbl["renumber"] + os.path.splitext(
instr)[1]
outstr = outpathgen(outdir,
subdir=finalid,
filename=fout,
mksubdir=True)
newstructure.write_minimal_pdb(outstr)
strset2[finalid] = structure
found = True
if found is False:
logger.warning("Identifier '" + key +
"' not found in sequence input.")
logger.info('Done' + os.linesep)
logger.info('Cropping renumbered structure(s)...')
outseq = os.path.join(
outdir,
os.path.splitext(os.path.basename(inseq))[0] + infixlbl["croprenum"] +
os.path.splitext(os.path.basename(inseq))[1])
for key, S in gseqset.items():
newS = S.deepcopy()
if key in intervals:
if insprot is not None and minlen > 0.0:
newinterval = {}
for key2, monomer in S.imer.items():
cropped_seq = False
for key3 in monomer.chains:
if key3 in intervals[key]:
if insprot is not None and minlen > 0.0:
newinterval[key3] = intervals[key][key3].deepcopy()
newinterval[key3].tags[
'description'] += ' - Uniprot threshold'
newinterval[key3].subint = []
unilbl = ' uniprot chains included: '
for unicode, uniintervals in intervals[key][
key3].tags['uniprot'].items():
uniseq = uniprotset[unicode].imer['1']
if 100 * uniintervals.n_elements(
) / uniseq.length() >= minlen:
newinterval[key3] = newinterval[
key3].union(
intervals[key][key3].intersection(
uniintervals))
unilbl += unicode + '|'
if cropped_seq is False:
monomer = cop.crop_seq(monomer,
newinterval[key3],
targetlbl + unilbl,
terms=args.terminals)
cropped_seq = True
else:
if cropped_seq is False:
monomer = cop.crop_seq(monomer,
intervals[key][key3],
targetlbl,
terms=args.terminals)
cropped_seq = True
if newS.imer[key2] != monomer:
newS.imer[key2] = monomer.deepcopy()
else:
logger.warning(
'Chain-name ' + key + '_' + str(key3) +
' not found in database. Cropping not performed.')
monomer.update_cropsheader()
hf = '_' + key2 if args.individual is True else ''
ifx = infixlbl["croprenum"] if cropped_seq is True else ''
fout = (key + hf + ifx +
os.path.splitext(os.path.basename(inseq))[1])
outseq = outpathgen(outdir,
subdir=key,
filename=fout,
mksubdir=True)
monomer.dump(outseq)
if monomer.cropmap is not None:
fout = key + hf + infixlbl["croprenum"] + '.cropmap'
outmap = outpathgen(outdir, subdir=key, filename=fout)
monomer.dumpmap(outmap)
cropped_str = cop.crop_pdb(strset2[key], newS, original_id=True)
fout = key + infixlbl["crop"] + os.path.splitext(instr)[1]
outstr = outpathgen(outdir,
subdir=key,
filename=fout,
mksubdir=True)
cropped_str.write_minimal_pdb(outstr)
cropped_str2 = cop.crop_pdb(strset2[key], newS, original_id=False)
fout = key + infixlbl["croprenum"] + os.path.splitext(instr)[1]
outstr = outpathgen(outdir,
subdir=key,
filename=fout,
mksubdir=True)
cropped_str2.write_minimal_pdb(outstr)
else:
logger.warning('PDB-ID ' + key.upper() +
' not found in database. Cropping not performed.')
for key2, monomer in newS.imer.items():
hf = '_' + key2 if args.individual is True else ''
fout = key + hf + os.path.splitext(os.path.basename(inseq))[1]
outseq = outpathgen(outdir,
subdir=key,
filename=fout,
mksubdir=True)
monomer.dump(outseq)
# FINISH
logger.info('Done' + os.linesep)
return
def main():
"""Renumber a structure file in agreement with the residue positions in the sequence file corresponding to that structure.
Non-polymer elements are numbered starting right after the final (TER) residue.
IMPORTANT: If the input sequence and the input structure files are not from the
same source (e.g. RCSB PDB) a source conflict might occur making the
renumbering operation unsuccessful even if the program does not crash.
"""
# INITIALISE AND PARSE ARGUMENTS FROM COMMAND LINE
parser = create_argument_parser()
args = parser.parse_args()
global logger
logger = ccl.crops_logger(level="info")
logger.info(ccl._welcome())
inseq = check_path(args.input_seqpath[0], 'file')
instr = check_path(args.input_strpath[0])
if args.outdir is None:
outdir = check_path(os.path.dirname(inseq), 'dir')
else:
outdir = check_path(os.path.join(args.outdir[0], ''), 'dir')
infixlbl = ".crops.seq"
# PARSE INPUT FILES
logger.info('Parsing sequence file ' + inseq)
if args.preselect is not None:
subset = set(args.preselect)
else:
subset = None
seqset = cin.parseseqfile(seq_input=inseq, inset=subset)
logger.info('Done')
logger.info('Parsing structure file ' + instr)
strset, fileset = cin.parsestrfile(instr)
logger.info('Done')
# MAIN OPERATION / PRINT OUT RESULTS WITHIN
logger.info('Renumbering structure(s)...')
for pdbid, structure in strset.items():
found = False
for seqname in seqset:
if ((seqname in pdbid) or (len(seqset) == 1 and len(strset) == 1)):
finalid = seqname
try:
newstructure = cop.renumber_pdb(seqset[seqname], structure)
except (AttributeError, IndexError) as e:
logger.warning(
'Something has gone wrong during renumbering:\n{}'.
format(e))
if args.force_alignment:
logger.info('Attempting Needleman-Wunsch...')
newstructure = cop.renumber_pdb_needleman(
seqset[seqname], structure)
else:
logger.critical(
'Unable to renumber the structure, exiting now. '
'Try again with -f option to force the alignment.')
return
fout = finalid + infixlbl + os.path.splitext(instr)[1]
outstr = outpathgen(outdir,
subdir=finalid,
filename=fout,
mksubdir=True)
newstructure.write_minimal_pdb(outstr)
found = True
if found is False:
logger.warning("Identifier '" + pdbid +
"' not found in sequence input.")
# FINISH
logger.info('Done' + os.linesep)
return
def main():
parser = create_argument_parser()
args = parser.parse_args()
global logger
logger = pcl.pisacov_logger(level="info")
welcomemsg, starttime = pcl.welcome(command=__script__)
logger.info(welcomemsg)
# PARSE CONFIGURATION FILE:
invals = pco._initialise_inputs()
invals['INSEQ'] = None
invals['INSTR'] = None
invals['ALTDB'] = None
invals['OUTROOT'] = None
invals['OUTCSVPATH'] = None
invals['UPTHRESHOLD'] = None
# READ INPUT ARGUMENTS
invals['INSEQ'] = ppaths.check_path(args.seqpath[0], 'file')
invals['INSTR'] = ppaths.check_path(args.crystalpath[0], 'file')
if args.hhblits_arguments is not None:
invals['HHBLITS_PARAMETERS'] = pco._check_hhparams(
args.hhblits_arguments)
else:
pass
if args.uniprot_threshold is not None:
try:
invals['UPTHRESHOLD'] = float(args.uniprot_threshold[0])
except ValueError:
logger.critical('Uniprot threshold given not valid.')
if invals['UNICLUST_FASTA_PATH'] is None:
invals['UNICLUST_FASTA_PATH'] = pco._uniurl
else:
pass
if args.skip_conpred is True:
skipexec = True
if (args.hhblits_arguments is not None
or args.uniprot_threshold is not None):
logger.info(
'HHblits, UniProt threshold parameters given bypassed by --skip_conpred'
)
else:
skipexec = False
cropping = args.remove_insertions
scoring = [cropping, not cropping]
if args.outdir is None:
invals['OUTROOT'] = ppaths.check_path(os.path.dirname(invals['INSEQ']))
else:
invals['OUTROOT'] = ppaths.check_path(os.path.join(args.outdir[0], ''))
ppaths.mdir(invals['OUTROOT'])
invals['OUTCSVPATH'] = []
if args.collection_file is None:
invals['OUTCSVPATH'].append(
ppaths.check_path(
os.path.join(invals['OUTROOT'],
("evcovsignal" + os.extsep + "cropped" +
os.extsep + "pisacov" + os.extsep + "csv"))))
invals['OUTCSVPATH'].append(
ppaths.check_path(
os.path.join(invals['OUTROOT'],
("evcovsignal" + os.extsep + "full" + os.extsep +
"pisacov" + os.extsep + "csv"))))
else:
if cropping is True:
invals['OUTCSVPATH'].append(
ppaths.check_path(args.collection_file[0]))
invals['OUTCSVPATH'].append(
ppaths.check_path(
os.path.splitext(args.collection_file[0])[0] + os.extsep +
'full' + os.extsep +
os.path.splitext(args.collection_file[0])[1]))
else:
invals['OUTCSVPATH'].append(None)
invals['OUTCSVPATH'].append(
ppaths.check_path(args.collection_file[0]))
# Define formats used
sources = pco._sources()
# Parse sequence and structure files
logger.info('Parsing sequence file...')
seqs = cps.parseseqfile(invals['INSEQ'])
logger.info('Parsing structure file...')
strs, filestrs = cps.parsestrfile(invals['INSTR'])
if len(seqs) == 1 or len(strs) == 1:
if len(seqs) == 1:
for key in seqs:
pdbid = key
elif len(seqs) > 1 and len(strs) == 1:
for key in strs:
for key2 in seqs:
if key.upper() == key2.upper():
pdbid = key.upper()
else:
if key2.upper() in key.upper():
pdbid = key2.upper()
else:
raise Exception(
'More than one pdbid in sequence and/or structure set.')
seq = seqs[pdbid]
#structure = strs[pdbid]
# CROPPING AND RENUMBERING
outpdbdir = os.path.join(invals['OUTROOT'], pdbid, "")
instrc = os.path.join(invals['OUTROOT'], pdbid,
os.path.basename(invals['INSTR']))
fseq = {}
fmsa = {}
if skipexec is False:
if cropping is True:
logger.info('Cropping and renumbering sequences, ' +
'structures according to SIFTS database.')
logger.info(pcl.running('CROPS-cropstr'))
itime = datetime.datetime.now()
psc.runcrops(invals['INSEQ'], invals['INSTR'],
invals['SIFTS_PATH'], invals['UPTHRESHOLD'],
invals['UNICLUST_FASTA_PATH'], invals['OUTROOT'])
logger.info(pcl.running('CROPS-cropstr', done=itime))
else:
logger.info('Renumbering structure ' +
'according to position in sequence.')
logger.info(pcl.running('CROPS-renumber'))
itime = datetime.datetime.now()
psc.renumcrops(invals['INSEQ'], invals['INSTR'], invals['OUTROOT'])
logger.info(pcl.running('CROPS-renumber', done=itime))
ppaths.mdir(outpdbdir)
if cropping is False:
psc.splitseqs(invals['INSEQ'], outpdbdir)
copyfile(invals['INSTR'], instrc)
for i, iseq in seq.imer.items():
fiseq = pdbid + '_' + i + '.fasta'
fseq[i] = os.path.join(invals['OUTROOT'], pdbid, fiseq)
fiseq = pdbid + '_' + i + '.msa.aln'
fmsa[i] = os.path.join(invals['OUTROOT'], pdbid, 'hhblits', fiseq)
if skipexec is False:
iseq.dump(fseq[i])
# Parse cropped sequences and maps
if cropping is True:
amap = {}
fcropseq = {}
fcropmsa = {}
for i, iseq in seq.imer.items():
fprefix = pdbid + '_' + i + '.crops.to_uniprot'
fmap = os.path.join(invals['OUTROOT'], pdbid,
fprefix + os.extsep + 'cropmap')
amap.update(cps.parsemapfile(fmap)[pdbid])
fcropseq[i] = os.path.join(invals['OUTROOT'], pdbid,
fprefix + os.extsep + 'fasta')
fcropmsa[i] = os.path.join(
invals['OUTROOT'], pdbid, 'hhblits',
(fprefix + os.extsep + 'msa' + os.extsep + 'aln'))
seq.set_cropmaps(amap, cropmain=True)
# EXECUTION OF EXTERNAL PROGRAMS
hhdir = os.path.join(invals['OUTROOT'], pdbid, 'hhblits', '')
dmpdir = os.path.join(invals['OUTROOT'], pdbid, 'dmp', '')
pisadir = os.path.join(invals['OUTROOT'], pdbid, 'pisa', '')
fstr = os.path.join(
invals['OUTROOT'],
(pdbid + os.extsep + 'crops' + os.extsep + 'seq' + os.extsep + 'pdb'))
if cropping:
fcropstr = os.path.join(
invals['OUTROOT'], pdbid,
(pdbid + os.extsep + 'crops' + os.extsep + 'oldids' + os.extsep +
'to_uniprot' + os.path.splitext(invals['INSTR'])[1]))
if skipexec is False:
# MSA GENERATOR
ppaths.mdir(hhdir)
if invals['HHBLITS_PARAMETERS'] == ['3', '0.001', 'inf', '50', '99']:
logger.info(
'Generating Multiple Sequence Alignment using DeepMetaPSICOV default parameters... [AS RECOMMENDED]'
)
elif invals['HHBLITS_PARAMETERS'] == ['2', '0.001', '1000', '0', '90']:
logger.info(
'Generating Multiple Sequence Alignment using HHBlits default parameters...'
)
else:
logger.info(
'Generating Multiple Sequence Alignment using user-custom parameters...'
)
for i, iseq in seq.imer.items():
sfile = fcropseq[i] if cropping is True else fseq[i]
afile = fcropmsa[i] if cropping is True else fmsa[i]
logger.info(pcl.running('HHBlits'))
itime = datetime.datetime.now()
themsa = psm.runhhblits(sfile, invals['HHBLITS_PARAMETERS'], hhdir)
logger.info(pcl.running('HHBlits', done=itime))
if cropping is True:
iseq.cropmsa = themsa
if iseq.ncrops() == 0:
iseq.msa = iseq.cropmsa
logger.info(' Cropped sequence ' + iseq.oligomer_id +
'_' + iseq.name +
' is identical to original sequence.')
continue
else:
pass
else:
iseq.msa = themsa
# DEEP META PSICOV RUN
ppaths.mdir(dmpdir)
if skipexec is False:
logger.info(
'Generating contact prediction lists via DeepMetaPSICOV...')
for i, iseq in seq.imer.items():
sfile = fcropseq[i] if cropping is True else fseq[i]
afile = fcropmsa[i] if cropping is True else fmsa[i]
nsfile = os.path.join(dmpdir, os.path.basename(sfile))
if sfile != nsfile:
copyfile(sfile, nsfile)
logger.info(pcl.running('DeepMetaPSICOV'))
itime = datetime.datetime.now()
psd.rundmp(nsfile, afile, dmpdir)
logger.info(pcl.running('DeepMetaPSICOV', done=itime))
# INTERFACE GENERATION, PISA
ppaths.mdir(pisadir)
if skipexec is False:
logger.info('Generating interface files via PISA...')
sfile = fcropstr if cropping is True else fstr
logger.info(pcl.running('PISA'))
itime = datetime.datetime.now()
iflist = psp.runpisa(sfile, pisadir, sessionid=pdbid)
logger.info(pcl.running('PISA', done=itime))
endmsg = pcl.ok(starttime, command=__script__)
logger.info(endmsg)
return