def main(args):
tags = {}
if args.verbose:
sys.stderr.write(">> reading in tag sequences...\n")
with nopen(args.tags) as fasta:
for name, seq in read_fasta(fasta):
tags[name] = seq
i = 0
for fx in args.reads:
if args.verbose:
sys.stderr.write(">> processing %s...\n" % op.basename(fx))
# process either fasta or fastq.
if ".fasta" in fx or ".fa" in fx:
with nopen(fx) as fa:
for f_id, f_seq in read_fasta(fa):
i += 1
if i % 1000000 == 0 and args.verbose:
sys.stderr.write(">> processed %d reads...\n" % i)
print_record(tags, f_id, f_seq)
else:
with nopen(fx) as fq:
for f_id, f_seq, f_qual in read_fastq(fq):
i += 1
if i % 1000000 == 0 and args.verbose:
sys.stderr.write(">> processed %d reads...\n" % i)
print_record(tags, f_id, f_seq)
def main(args):
# fields from issake
fields = "contig_id length reads avg_coverage seed v_region j_region".split()
# the only fields i believe make any sense to keep
out = "id v_region j_region length reads avg_coverage percent_of_total sequence".split()
# total reads used in assembly
total = 0.
with nopen(args.fasta_in) as fasta:
for name, seq in read_fasta(fasta):
name = name.replace("size","").replace("cov","").replace("read","").replace("seed:","")
d = dict(zip(fields, name.split("|")))
total += int(d['reads'])
with nopen(args.fasta_in) as fasta,\
open(args.fasta_out, 'wb') as fasta_out,\
open(args.meta, 'wb') as meta:
# print header
meta.write("\t".join(out) + "\n")
for i, (name, seq) in enumerate(read_fasta(fasta)):
# remove some text from iSSAKE output
name = name.replace("size","").replace("cov","").replace("read","").replace("seed:","")
d = dict(zip(fields, name.split("|")))
# want to shorten the read names
d['id'] = "contig_%d" % i
d['percent_of_total'] = "%.6g" % (100 * (int(d['reads']) / total))
d['sequence'] = seq
meta.write("\t".join(map(str, [d[o] for o in out])) + "\n")
write_fasta(fasta_out, d['id'], seq.upper())
def main(args):
full_seqs = {}
tags = {}
with nopen(args.fasta) as fasta:
for name, seq in read_fasta(fasta):
full_seqs[name] = seq.upper()
# each tcr in original fasta
for tcr, seq in full_seqs.iteritems():
unique_expected = len(full_seqs) - 1
# all possible tags
for i in range(len(seq) - args.length, 0, -1):
# tag matches favor 3' end
tag = seq[i:args.length + i]
# reached the end of the sequence
if len(tag) < args.length: break
unique_found = 0
# tag not present in any other tcr
for ss_tcr, ss_seq in full_seqs.iteritems():
# the current tcr
if ss_tcr == tcr: continue
# finding unique tags
if ss_seq.find(tag) == -1:
unique_found += 1
if unique_found == unique_expected:
tags[tcr] = tag
# exit loop on first unique tag
break
# ensure this actually worked
taglist = [tag for name, tag in tags.iteritems()]
tagset = set(taglist)
assert(len(taglist) == len(tagset))
taglist = []
# print results
for tcr, tag in tags.iteritems():
taglist.append(tcr)
print ">%s\n%s" % (tcr, tag)
# tags found stats
if args.verbose:
sys.stderr.write("Of %d regions, %d tags were found.\n" \
% (len(full_seqs), len(tagset)))
alltcrs = []
for tcr, seq in full_seqs.iteritems():
alltcrs.append(tcr)
alltcrs = set(alltcrs)
taglist = set(taglist)
diff = alltcrs - taglist
if diff:
sys.stderr.write("Unable to find a unique tag for:\n")
sys.stderr.write("\n".join(diff) + "\n")
#!/usr/bin/env python
# encoding: utf-8
"""
Parses the read name down to only include the necessary gene label.
"""
import re
import sys
import itertools
from toolshed import nopen
from parsers import read_fasta
def main(args):
with nopen(args.fasta) as fasta
for name, seq in read_fasta(fasta):
try:
# rename from imgt
name = re.findall(r'(%s[^\|]+)' % args.gene.upper(), name)[0]
print ">%s\n%s" % (name, seq.upper())
except IndexError:
sys.stderr.write(">> unable to parse: %s\n>> for gene: %s\n" \
% (name, args.gene))
pass
if __name__ == "__main__":
import argparse
p = argparse.ArgumentParser(description=__doc__,
formatter_class=argparse.RawDescriptionHelpFormatter)
p.add_argument('fasta')
req = p.add_argument_group('required arguments')
req.add_argument('-g', '--gene', required=True,
help="gene name, eg. TRAJ or TRBV")
