def main():
parser = argparse.ArgumentParser(
description='Use reference junctions when they are close',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('--min_intron_size',
type=int,
default=68,
help="INT min intron size")
parser.add_argument(
'--min_local_support',
type=int,
default=0,
help=
"INT min number of junctions within search_size of a junction in order to count it"
)
parser.add_argument('--search_size',
type=int,
default=10,
help="INT search space for reference")
parser.add_argument(
'--output_fake_psl',
help="FASTAFILE reference genome to make a fake PSL output")
parser.add_argument('psl', help="PSLFILENAME or '-' for STDIN")
parser.add_argument('reference_genepred',
help="FASTAFILENAME for reference genepred")
args = parser.parse_args()
cpus = multiprocessing.cpu_count()
genome = {}
if args.output_fake_psl:
genome = read_fasta_into_hash(args.output_fake_psl)
#read in the reference genepred first
gpf = GenePredBasics.GenePredFile(args.reference_genepred)
#lets sort entries by chromosome
ref = {}
for e in [x.entry for x in gpf.entries]:
if len(e['exonStarts']) <= 1: continue
if e['chrom'] not in ref:
ref[e['chrom']] = {}
for i in range(1, len(e['exonStarts'])):
if e['exonEnds'][i - 1] not in ref[e['chrom']]:
ref[e['chrom']][e['exonEnds'][i - 1]] = {}
if e['exonStarts'][i] + 1 not in ref[e['chrom']][e['exonEnds'][i -
1]]:
ref[e['chrom']][e['exonEnds'][i - 1]][e['exonStarts'][i] +
1] = e['strand']
#Stored all junctions as 1-base
read_info = {}
pf = GenericFileReader(args.psl)
fcount_total = 0
while True:
line = pf.readline()
if not line: break
if re.match('^#', line): continue
line = line.rstrip()
pe = PSLBasics.line_to_entry(line)
if len(pe['tStarts']) != len(pe['blockSizes']) or len(
pe['qStarts']) != len(pe['blockSizes']):
sys.stderr.write("WARNING invalid psl\n")
continue
genepred_line = PSLBasics.convert_entry_to_genepred_line(pe)
ge = GenePredBasics.smooth_gaps(
GenePredBasics.line_to_entry(genepred_line), args.min_intron_size)
refjuns = {}
if pe['tName'] in ref: refjuns = ref[pe['tName']]
new_ge = nudge(pe, ge, refjuns, args)
if args.output_fake_psl:
new_psl_line = GenePredBasics.entry_to_fake_psl_line(
new_ge, genome)
print new_psl_line
else:
print GenePredBasics.entry_to_line(new_ge)
def main():
parser = argparse.ArgumentParser()
parser.add_argument('a', nargs=1, help='FILENAME genepred file A')
parser.add_argument('b', nargs=1, help='FILENAME genepred file B')
#parser.add_argument('-p',nargs='?',help='INT the number of threads to run.')
parser.add_argument('--minexoncount',
nargs='?',
help='INT the minimum number of exons required.')
parser.add_argument(
'--minoverlap_internal',
nargs='?',
help=
'FLOAT the fraction (0-1) of the required reciprocal overlap of an internal exon to call an exon a match.'
)
parser.add_argument(
'--minoverlap_first',
nargs='?',
help=
'FLOAT the fraction (0-1) of the required reciprocal overlap of the first exon to call an exon a match.'
)
parser.add_argument(
'--minoverlap_last',
nargs='?',
help=
'FLOAT the fraction (0-1) of the required reciprocal overlap of the last exon to call an exon a match.'
)
parser.add_argument(
'--minoverlap',
nargs='?',
help=
'FLOAT the fraction (0-1) of the required reciprocal overlap of any exon to call an exon a match.'
)
parser.add_argument(
'--leftouterjoin',
action='store_true',
help=
'Output the entry A regardless of whether a matching entry in B is found'
)
parser.add_argument('--output_a_not_in_b',
action='store_true',
help='Output entries that occur in A but not B')
parser.add_argument(
'--best_b_only',
action='store_true',
help=
'Output only one entry of B for each A and try to pick the best based on reciprocal overlap'
)
parser.add_argument(
'--allow_a_subset_of_b_fragments',
action='store_true',
help=
'If A is just a subset of B, then call it as a match. This means all exons of A found a conecutive match, but B could have more exons on either end.'
)
parser.add_argument(
'--allow_any_fragments',
action='store_true',
help='If set, allow any partial match, not just the best')
args = parser.parse_args()
#pcount = multiprocessing.cpu_count()
#if args.p: pcount = int(args.p)
# go through contingencies of overlap requirements and set them
overlap = [0, 0, 0]
if args.minoverlap:
overlap = [
float(args.minoverlap),
float(args.minoverlap),
float(args.minoverlap)
]
if args.minoverlap_first:
overlap[0] = float(args.minoverlap_last)
if args.minoverlap_last:
overlap[2] = float(args.minoverlap_last)
if args.minoverlap_internal:
overlap[1] = float(args.minoverlap_internal)
# read the genepred files
gpdA = GenePredBasics.GenePredFile(args.a[0])
gpdB = GenePredBasics.GenePredFile(args.b[0])
#if pcount > 1:
# p = multiprocessing.Pool(processes=pcount)
for eA in gpdA.entries:
#if pcount > 1:
# p.apply_async(check_B_entries,[eA,overlap,args])
#else:
check_B_entries(eA, gpdB, overlap, args)