def extract_reads_for_lane(fastq,lane):
"""
Fetch reads from Fastq from specified lane
Generator function which iterates through a
Fastqe file and yields each read record where
the lane number matches the specified lane.
Example usage:
>>> for r in extract_reads_for_lane('illumina_R1.fq',2):
>>> ... print r
Arguments:
fastq (str): path to Fastq (can be gzipped)
Yields:
String: matching read record as a string.
"""
regex_pattern = r"^([^:]*:){3}%s:" % lane
for read in getreads_regex(fastq,regex_pattern):
yield '\n'.join(read)
def extract_reads_for_lane(fastq, lane):
"""
Fetch reads from Fastq from specified lane
Generator function which iterates through a
Fastqe file and yields each read record where
the lane number matches the specified lane.
Example usage:
>>> for r in extract_reads_for_lane('illumina_R1.fq',2):
>>> ... print(r)
Arguments:
fastq (str): path to Fastq (can be gzipped)
Yields:
String: matching read record as a string.
"""
regex_pattern = r"^([^:]*:){3}%s:" % lane
for read in getreads_regex(fastq, regex_pattern):
yield '\n'.join(read)
def main(args=None):
# Command line processing
if args is None:
args = sys.argv[1:]
p = optparse.OptionParser(usage="%prog -m PATTERN |-n NREADS infile "
"[ infile ... ]",
version="%prog "+__version__,
description=__description__)
p.add_option('-m','--match',action='store',dest='pattern',default=None,
help="extract records that match Python regular "
"expression PATTERN")
p.add_option('-n',action='store',dest='n',default=None,
help="extract N random reads from the input file(s). "
"If multiple files are supplied (e.g. R1/R2 pair) then "
"the same subsets will be extracted for each. "
"(Optionally a percentage can be supplied instead e.g. "
"'50%' to extract a subset of half the reads.)")
p.add_option('-s','--seed',action='store',dest='seed',default=None,
help="specify seed for random number generator (used "
"for -n option; using the same seed should produce the "
"same 'random' sample of reads)")
opts,args = p.parse_args(args)
if len(args) < 1:
p.error("Need to supply at least one input file")
# Pattern matching option
if opts.pattern is not None:
if opts.n is not None:
p.error("Need to supply only one of -n or -m options")
print "Extracting reads matching '%s'" % opts.pattern
for f in args:
if f.endswith('.gz'):
outfile = os.path.basename(os.path.splitext(f[:-3])[0])
else:
outfile = os.path.basename(os.path.splitext(f)[0])
outfile += '.subset_regex.fq'
print "Extracting to %s" % outfile
with open(outfile,'w') as fp:
for read in getreads_regex(f,opts.pattern):
fp.write('\n'.join(read) + '\n')
else:
# Seed random number generator
if opts.seed is not None:
random.seed(opts.seed)
# Count the reads
nreads = sum(1 for i in getreads(args[0]))
print "Number of reads: %s" % nreads
if len(args) > 1:
print "Verifying read numbers match between files"
for f in args[1:]:
if sum(1 for i in getreads(f)) != nreads:
print "Inconsistent numbers of reads between files"
sys.exit(1)
# Generate a subset of read indices to extract
try:
nsubset = int(opts.n)
except ValueError:
if str(opts.n).endswith('%'):
nsubset = int(float(opts.n[:-1])*nreads/100.0)
if nsubset > nreads:
print "Requested subset (%s) is larger than file (%s)" % (nsubset,
nreads)
sys.exit(1)
print "Generating set of %s random indices" % nsubset
subset_indices = random.sample(xrange(nreads),nsubset)
# Extract the reads to separate files
for f in args:
if f.endswith('.gz'):
outfile = os.path.basename(os.path.splitext(f[:-3])[0])
else:
outfile = os.path.basename(os.path.splitext(f)[0])
outfile += '.subset_%s.fq' % nsubset
print "Extracting to %s" % outfile
with open(outfile,'w') as fp:
for read in getreads_subset(f,subset_indices):
fp.write('\n'.join(read) + '\n')
def main(args=None):
# Command line processing
if args is None:
args = sys.argv[1:]
p = argparse.ArgumentParser(version="%(prog)s "+__version__,
description=__description__)
p.add_argument('-m','--match',action='store',dest='pattern',
default=None,
help="extract records that match Python regular "
"expression PATTERN")
p.add_argument('-n',action='store',dest='n',default=None,
help="extract N random reads from the input file(s). "
"If multiple files are supplied (e.g. R1/R2 pair) then "
"the same subsets will be extracted for each. "
"(Optionally a percentage can be supplied instead e.g. "
"'50%%' to extract a subset of half the reads.)")
p.add_argument('-s','--seed',action='store',dest='seed',default=None,
help="specify seed for random number generator (used "
"for -n option; using the same seed should produce the "
"same 'random' sample of reads)")
p.add_argument('infiles',metavar='infile',nargs='+',
help="input FASTQ, CSFASTA, or QUAL file")
args = p.parse_args(args)
# Pattern matching option
if args.pattern is not None:
if args.n is not None:
p.error("Need to supply only one of -n or -m options")
print("Extracting reads matching '%s'" % args.pattern)
for f in args.infiles:
if f.endswith('.gz'):
outfile = os.path.basename(os.path.splitext(f[:-3])[0])
else:
outfile = os.path.basename(os.path.splitext(f)[0])
outfile += '.subset_regex.fq'
print("Extracting to %s" % outfile)
with open(outfile,'w') as fp:
for read in getreads_regex(f,args.pattern):
fp.write('\n'.join(read) + '\n')
else:
# Seed random number generator
if args.seed is not None:
random.seed(args.seed)
# Count the reads
nreads = sum(1 for i in getreads(args.infiles[0]))
print("Number of reads: %s" % nreads)
if len(args.infiles) > 1:
print("Verifying read numbers match between files")
for f in args.infiles[1:]:
if sum(1 for i in getreads(f)) != nreads:
print("Inconsistent numbers of reads between files")
sys.exit(1)
# Generate a subset of read indices to extract
try:
nsubset = int(args.n)
except ValueError:
if str(args.n).endswith('%'):
nsubset = int(float(args.n[:-1])*nreads/100.0)
if nsubset > nreads:
print("Requested subset (%s) is larger than file (%s)" % (nsubset,
nreads))
sys.exit(1)
print("Generating set of %s random indices" % nsubset)
subset_indices = random.sample(xrange(nreads),nsubset)
# Extract the reads to separate files
for f in args.infiles:
if f.endswith('.gz'):
outfile = os.path.basename(os.path.splitext(f[:-3])[0])
else:
outfile = os.path.basename(os.path.splitext(f)[0])
outfile += '.subset_%s.fq' % nsubset
print("Extracting to %s" % outfile)
with open(outfile,'w') as fp:
for read in getreads_subset(f,subset_indices):
fp.write('\n'.join(read) + '\n')