def main(gff_file=None, fasta_file=None, stype=None, dline=None, qc=True, output_prefix=None, logger=None):
if logger == None:
logger = logging.getLogger(__name__+'stderr')
logger.setLevel(logging.INFO)
stderr_handler = logging.StreamHandler()
stderr_handler.setFormatter(logging.Formatter('%(levelname)-8s %(message)s'))
logger.addHandler(stderr_handler)
logger_null = logging.getLogger(__name__+'null')
null_handler = logging.NullHandler()
logger_null.addHandler(null_handler)
if output_prefix:
logger.info('Specifying prefix of output file name: (%s)...', output_prefix)
fname = '{0:s}_{1:s}.fa'.format(output_prefix, stype)
report_fh = open(fname, 'wb')
else:
parser.print_help()
sys.exit(1)
if not gff_file or not fasta_file or not stype:
print('All of Gff file, fasta file, and type of extracted seuqences need to be specified')
return
type_set=['gene','exon','pre_trans', 'trans', 'cds', 'pep']
if not stype in type_set:
logger.error('Your sequence type is "{0:s}". Sequence type must be one of {1:s}!'.format(stype, str(type_set)))
return
logger.info('Reading files: {0:s}, {1:s}...'.format(gff_file, fasta_file))
gff = Gff3(gff_file=gff_file, fasta_external=fasta_file, logger=logger)
if qc:
logger.info('Checking errors...')
gff.check_parent_boundary()
gff.check_phase()
gff.check_reference()
error_set = function4gff.extract_internal_detected_errors(gff)
t = intra_model.main(gff, logger=logger)
if t:
error_set.extend(t)
t = single_feature.main(gff, logger=logger)
if t:
error_set.extend(t)
if len(error_set):
escaped_error = ['Esf0012','Esf0033']
eSet = list()
for e in error_set:
if not e['eCode'] in escaped_error:
eSet.append(e)
if len(eSet):
logger.warning('The extracted sequences might be wrong for the following features which have formatting errors...')
print('ID\tError_Code\tError_Tag')
for e in eSet:
tag = '[{0:s}]'.format(e['eTag'])
print e['ID'], e['eCode'], tag
logger.info('Extract seqeunces for {0:s}...'.format(stype))
seq=dict()
if stype == 'pre_trans' or stype == 'gene' or stype == 'exon':
seq = extract_start_end(gff, stype, dline)
elif stype == 'trans':
feature_type = ['exon', 'pseudogenic_exon']
seq = splicer(gff, feature_type, dline)
elif stype == 'cds':
feature_type = ['CDS']
seq = splicer(gff, feature_type, dline)
elif stype == 'pep':
feature_type = ['CDS']
tmpseq = splicer(gff, feature_type, dline)
for k,v in tmpseq.items():
k = k.replace("|mRNA(CDS)|", "|peptide|").replace("-RA", "-PA")
v = translator(v)
seq[k] = v
if len(seq):
logger.info('Print out extracted sequences: {0:s}_{1:s}.fa...'.format(output_prefix, stype))
for k,v in seq.items():
report_fh.write('{0:s}\n{1:s}\n'.format(k,v))
#ERROR_CODE = ['Esf0001', 'Esf0002', 'Ema0005', 'Emr0001']
#ERROR_TAG = ['pseudogene or not?', 'Negative/Zero start/end coordinate', 'unusual child features in the type of pseudogene found', 'Duplicate transcripts found']
#ERROR_INFO = dict(zip(ERROR_CODE, ERROR_TAG))
logger_stderr.info('Reading gff files: (%s)...\n', args.gff)
gff3 = Gff3(gff_file=args.gff, fasta_external=args.fasta, logger=logger_null)
gff3.check_unresolved_parents()
gff3.check_parent_boundary()
gff3.check_phase()
gff3.check_reference()
logger_stderr.info('Checking missing attributes: (%s)...\n', 'single_feature.FIX_MISSING_ATTR()')
error_set = list()
if function4gff.extract_internal_detected_errors(gff3):
error_set.extend(function4gff.extract_internal_detected_errors(gff3))
if intra_model.main(gff3, logger=logger_stderr):
error_set.extend(intra_model.main(gff3, logger=logger_stderr))
if inter_model.main(gff3, logger=logger_stderr):
error_set.extend(inter_model.main(gff3, logger=logger_stderr))
if inter_model.main(gff3, logger=logger_stderr):
error_set.extend(single_feature.main(gff3, logger=logger_stderr))
if args.output:
logger_stderr.info('Print QC report at {0:s}'.format(args.output))
else:
logger_stderr.info('Print QC report at {0:s}'.format('report.txt'))
report_fh.write('ID\tError_code\tError_tag\n')
for e in error_set:
tag = '[{0:s}]'.format(e['eTag'])
report_fh.write('{0:s}\t{1:s}\t{2:s}\n'.format(str(e['ID']), str(e['eCode']), str(tag)))