def make_exons(args, thread_index, thread_count):
is_sam = True
if re.search('\.bam$', args.sam_file):
is_sam = False
stag = ''
if is_sam: stag = '-S'
cmd = 'samtools view -F 4 ' + stag + ' ' + args.sam_file
spcf = SamBasics.SAMtoPSLconversionFactory()
if args.reference_genome:
spcf.set_genome(args.reference_genome)
sampipe = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE)
fname = args.tempdir + '/bedpart.' + str(thread_index) + '.bed'
of = open(fname, 'w')
z = 0
with sampipe.stdout as inf:
for line in inf:
z += 1
if z % thread_count != thread_index: continue
line = line.rstrip()
if SamBasics.is_header(line):
continue
d = SamBasics.sam_line_to_dictionary(line)
strand = '+'
if SamBasics.check_flag(d['flag'], 16):
strand = '-'
seqs = []
sequence = d['seq']
seqs.append([d['qname'], d['rname'], strand, d['pos'], d['cigar']])
m = re.search('XA:Z:(\S+)', line)
if m and args.use_secondary_alignments:
e = m.group(1)
secondaries = e.rstrip(";").split(";")
for secondary in secondaries:
m1 = re.match('([^,]+),([+-])(\d+),([^,]+)', secondary)
if not m1:
sys.stderr.write("strange secondary format " +
secondary + "\n")
sys.exit()
seqs.append([
d['qname'],
m1.group(1),
m1.group(2),
int(m1.group(3)),
m1.group(4)
])
#p.apply_async(get_exons_from_seqs,[seqs,d,spcf])
exons = get_exons_from_seqs(seqs, d, spcf)
of.write(exons)
#return exons
of.close()
def main():
parser = argparse.ArgumentParser(
description=
"Find mapping distance of paired end reads. Takes an ordered (by query) alignment to a transcriptome.\nSomething that works for an input thus far is like:\nhisat --reorder -x mytranscriptome -1 my_1.fastq -2 my_2.fastq | this_script.py -"
)
parser.add_argument(
'input_sam',
help="SAMFILE ordered alignment a transcriptome or - for stdin")
args = parser.parse_args()
inf = sys.stdin
if args.input_sam != '-':
inf = open(args.input_sam)
msr = SamBasics.MultiEntrySamReader(inf)
spcf = SamBasics.SAMtoPSLconversionFactory()
data = []
sys.stderr.write("Pairs Mean Stddev\n")
while True:
entries = msr.read_entries()
if not entries: break
if len(entries) != 2: continue
[e1, e2] = entries
if e1.check_flag(4) or e2.check_flag(4): continue
if not e1.check_flag(2) and e2.check_flag(2): continue
if not ((e1.check_flag(64) and e2.check_flag(128)) or
(e1.check_flag(128) and e2.check_flag(64))):
continue
p1 = spcf.convert_line(e1.get_line())
p2 = spcf.convert_line(e2.get_line())
if not p1 or not p2: continue
p1 = PSLBasics.PSL(p1)
p2 = PSLBasics.PSL(p2)
dist = max(
p2.value('tEnd') - p1.value('tStart'),
p1.value('tEnd') - p2.value('tStart'))
data.append(dist)
if len(data) < 2: continue
if len(data) % 1000 == 0:
sys.stderr.write(
str(len(data)) + " " + str(int(mean(data))) + " " +
str(int(stddev(data))) + " \r")
sys.stderr.write(
str(len(data)) + " " + str(int(mean(data))) + " " +
str(int(stddev(data))) + " \r")
sys.stderr.write("\n")
def main():
parser = argparse.ArgumentParser(
description="Convert a sam file into a psl file")
parser.add_argument('--genome',
help="FASTA input file of reference genome")
parser.add_argument('--get_secondary_alignments',
action='store_true',
help="Report SA:Z secondary alignments as well")
parser.add_argument('--get_alternative_alignments',
action='store_true',
help="Report XA:Z alternative alignments as well")
parser.add_argument(
'--get_all_alignments',
action='store_true',
help="Report SA:Z and XA:Z alternative alignments as well")
parser.add_argument('--give_unique_names',
action='store_true',
help="Output query names will be unique.")
group = parser.add_mutually_exclusive_group()
group.add_argument(
'--output_fasta',
help=
"FILENAME to save an outgoing fasta. Only works for primary alignments."
)
group.add_argument(
'--output_fastq',
help=
"FILENAME to save an outgoing fastq. Only works for primary alignments."
)
parser.add_argument('infile', help="FILENAME input file or '-' for STDIN")
parser.add_argument('-o',
'--output',
help="FILENAME for the output, STDOUT if not set.")
args = parser.parse_args()
if (args.output_fasta
or args.output_fastq) and (args.get_secondary_alignments
or args.get_alternative_alignments
or args.get_all_alignments):
sys.stderr.write(
"ERROR, can only output the fastq/fasta if we are doing primary alignments only.\n"
)
sys.exit()
inf = sys.stdin
if args.infile != '-':
inf = open(args.infile)
of = sys.stdout
if args.output:
of = open(args.output, 'w')
spcf = SamBasics.SAMtoPSLconversionFactory()
if args.genome: spcf.set_genome(args.genome)
off = None
if args.output_fasta:
off = open(args.output_fasta, 'w')
if args.output_fastq:
off = open(args.output_fastq, 'w')
z = 0
for line in inf:
line = line.rstrip()
if SamBasics.is_header(line):
spcf.read_header_line(line)
continue
# We have a line to convert
psl = spcf.convert_line(line)
if psl:
pobj = PSL(psl)
z += 1
if args.give_unique_names:
pobj.entry['qName'] = 'Q' + str(z)
of.write(pobj.get_line() + "\n")
if args.output_fastq or args.output_fasta:
sam = SamBasics.SAM(line)
sequence = sam.value('seq').upper()
quality = sam.value('qual')
if sam.check_flag(16):
sequence = rc(sam.value('seq').upper())
quality = sam.value('qual')[::-1]
if args.output_fasta:
off.write(">" + pobj.value('qName') + "\n" + sequence +
"\n")
elif args.output_fastq:
if len(sequence) == len(quality):
off.write("@" + pobj.value('qName') + "\n" + sequence +
"\n" + "+\n" + quality + "\n")
else:
sys.stderr.write("ERROR: sequence " + sequence +
" length (" + str(len(sequence)) +
") doesnt match quality " + quality +
" length (" + str(len(quality)) +
")\n")
sys.exit()
# Lets look for secondary alignments to convert
if args.get_secondary_alignments or args.get_all_alignments:
secondary_alignments = SamBasics.get_secondary_alignments(
line.rstrip())
for samline in secondary_alignments:
psl = spcf.convert_line(samline)
if psl:
#print "\nsecondary"
#print samline
z += 1
pobj = PSL(psl)
if args.give_unique_names:
pobj.entry['qName'] = 'Q' + str(z)
of.write(pobj.get_line() + "\n")
if args.get_alternative_alignments or args.get_all_alignments:
alternative_alignments = SamBasics.get_alternative_alignments(
line.rstrip())
for samline in alternative_alignments:
psl = spcf.convert_line(samline)
if psl:
#print "\nsecondary"
#print samline
z += 1
pobj = PSL(psl)
if args.give_unique_names:
pobj.entry['qName'] = 'Q' + str(z)
of.write(pobj.get_line() + "\n")
inf.close()
of.close()