def run(self):
"""Subset reads based on read annotation and subset rules."""
infoMsg = "Extracting reads from {f} based on ".format(f=self.inFN)
infoMsg += "rules(FullLength={fl}, nonChimeric={nc}).".format(
fl="true" if self.rules.FL != 0 else "false",
nc="true" if self.rules.nonChimeric != 0 else "false")
logging.info(infoMsg)
if not self.printReadLengthOnly:
with FastaReader(self.inFN) as reader, \
FastaWriter(self.outFN) as writer:
for r in reader:
#print >> sys.stderr, r.name, self.ignore_polyA
annotation = ReadAnnotation.fromString(r.name,
self.ignore_polyA)
if self.satisfy(annotation, self.rules):
writer.writeRecord(r.name, r.sequence)
else: # print read length only, dont print read names and sequences
with FastaReader(self.inFN) as reader, \
open(self.outFN, 'w') as writer:
for r in reader:
annotation = ReadAnnotation.fromString(r.name,
self.ignore_polyA)
if self.satisfy(annotation, self.rules):
writer.write("{rl}\n".format(rl=len(r.sequence)))
def run(self):
"""Subset reads based on read annotation and subset rules."""
infoMsg = "Extracting reads from {f} based on ".format(f=self.inFN)
infoMsg += "rules(FullLength={fl}, nonChimeric={nc}).".format(
fl="true" if self.rules.FL != 0 else "false",
nc="true" if self.rules.nonChimeric != 0 else "false")
logging.info(infoMsg)
if not self.printReadLengthOnly:
with FastaReader(self.inFN) as reader, \
FastaWriter(self.outFN) as writer:
for r in reader:
#print >> sys.stderr, r.name, self.ignore_polyA
annotation = ReadAnnotation.fromString(
r.name, self.ignore_polyA)
if self.satisfy(annotation, self.rules):
writer.writeRecord(r.name, r.sequence)
else: # print read length only, dont print read names and sequences
with FastaReader(self.inFN) as reader, \
open(self.outFN, 'w') as writer:
for r in reader:
annotation = ReadAnnotation.fromString(
r.name, self.ignore_polyA)
if self.satisfy(annotation, self.rules):
writer.write("{rl}\n".format(rl=len(r.sequence)))
def test_satisfy(self):
"""Test function satisfy()."""
inFN = op.join(self.testDir, "data/test_subset.fa")
reads = []
with FastaReader(inFN) as reader:
reads = [x for x in reader]
rules = SubsetRules(1, 1) # Full-length, non-chimeric
obj = ReadsSubsetExtractor("in", "out", rules, True)
ans = [ReadAnnotation.fromString(r.name) for r in reads]
res = [obj.satisfy(an, rules) for an in ans]
expected = [1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 0, 1, 1]
self.assertTrue(res == expected)
def _updateChimeraInfo(self,
suspicous_hits,
in_read_fn,
out_nc_fn,
out_c_fn,
primer_report_fn,
write_report_header=True):
"""
in_read_fn --- a fasta of full-length reads or a fasta of
non-full-length reads.
For each full-length read in in_read_fn FASTA file, detect whether
it is chimeric or not, and write its annotation to
primer_report_fn.
Return:
(num_nc, num_c, num_nc_bases, num_c_bases)
"""
logging.debug(
"Update chimera info for reads in {f} ".format(f=in_read_fn))
logging.debug(
"Write primer report to {rpt}".format(rpt=primer_report_fn))
num_nc, num_c, num_nc_bases, num_c_bases = 0, 0, 0, 0
with FastaReader(in_read_fn) as reader, \
FastaWriter(out_nc_fn) as writer, \
FastaWriter(out_c_fn) as writer_chimera, \
open(primer_report_fn, 'w') as reporter:
if write_report_header:
reporter.write(ReadAnnotation.header(delimiter=",") + "\n")
for r in reader:
# e.g. r.name="movie/zmw/0_100_CCS fiveend=1;threeend=100;"
readid = r.name.split()[0]
annotation = ReadAnnotation.fromString(
r.name, ignore_polyA=self.ignore_polyA)
if readid not in suspicous_hits: # Non-chimeric reads
# Primer of a primer-trimmed read can not be None.
# assert(annotation.primer is not None)
annotation.chimera = 0
num_nc += 1
num_nc_bases += len(r.sequence)
writer.writeRecord(annotation.toAnnotation(), r.sequence)
else: # chimeric reads
annotation.chimera = 1
num_c += 1
num_c_bases += len(r.sequence)
writer_chimera.writeRecord(annotation.toAnnotation(),
r.sequence)
reporter.write(annotation.toReportRecord(delimitor=",") + "\n")
return (num_nc, num_c, num_nc_bases, num_c_bases)
def _updateChimeraInfo(self, suspicous_hits, in_read_fn, out_nc_fn,
out_c_fn, primer_report_fn,
write_report_header=True):
"""
in_read_fn --- a fasta of full-length reads or a fasta of
non-full-length reads.
For each full-length read in in_read_fn FASTA file, detect whether
it is chimeric or not, and write its annotation to
primer_report_fn.
Return:
(num_nc, num_c, num_nc_bases, num_c_bases)
"""
logging.debug("Update chimera info for reads in {f} ".
format(f=in_read_fn))
logging.debug("Write primer report to {rpt}".
format(rpt=primer_report_fn))
num_nc, num_c, num_nc_bases, num_c_bases = 0, 0, 0, 0
with FastaReader(in_read_fn) as reader, \
FastaWriter(out_nc_fn) as writer, \
FastaWriter(out_c_fn) as writer_chimera, \
open(primer_report_fn, 'w') as reporter:
if write_report_header:
reporter.write(ReadAnnotation.header(delimiter=",") + "\n")
for r in reader:
# e.g. r.name="movie/zmw/0_100_CCS fiveend=1;threeend=100;"
readid = r.name.split()[0]
annotation = ReadAnnotation.fromString(r.name,
ignore_polyA=self.ignore_polyA)
if readid not in suspicous_hits: # Non-chimeric reads
# Primer of a primer-trimmed read can not be None.
# assert(annotation.primer is not None)
annotation.chimera = 0
num_nc += 1
num_nc_bases += len(r.sequence)
writer.writeRecord(annotation.toAnnotation(),
r.sequence)
else: # chimeric reads
annotation.chimera = 1
num_c += 1
num_c_bases += len(r.sequence)
writer_chimera.writeRecord(annotation.toAnnotation(),
r.sequence)
reporter.write(annotation.toReportRecord(delimitor=",") + "\n")
return (num_nc, num_c, num_nc_bases, num_c_bases)
def _updateChimeraInfo(self, suspicous_hits, in_read_fn, out_flnc_fn,
out_flc_fn, primer_report_fl_fn):
"""
in_read_fn --- a fasta of full-length reads
For each full-length read in in_read_fn FASTA file, detect whether
it is chimeric or not, and write its annotation to
primer_report_fl_fn.
"""
logging.info("Update chimera info to reads annotations " +
"in the output FASTA file and the primer report.")
with FastaReader(in_read_fn) as reader, \
FastaWriter(out_flnc_fn) as writer, \
FastaWriter(out_flc_fn) as writer_chimera, \
open(primer_report_fl_fn, 'w') as reporter:
reporter.write("\t".join(ReadAnnotation.fieldsNames()) + "\n")
for r in reader:
# e.g. r.name="movie/zmw/0_100_CCS fiveend=1;threeend=100;"
readid = r.name.split()[0]
annotation = ReadAnnotation.fromString(r.name, ignore_polyA=self.ignore_polyA)
if readid not in suspicous_hits: # Non-chimeric reads
# Primer of a primer-trimmed read can not be None.
# assert(annotation.primer is not None)
annotation.chimera = 0
assert(annotation.isFullLength)
self.summary.num_flnc += 1
self.summary.num_flnc_bases += len(r.sequence)
writer.writeRecord(annotation.toAnnotation(),
r.sequence)
else: # chimeric reads
annotation.chimera = 1
self.summary.num_flc += 1
writer_chimera.writeRecord(annotation.toAnnotation(),
r.sequence)
reporter.write(annotation.toReportRecord() + "\n")
def _updateChimeraInfo(self, suspicous_hits, in_read_fn, out_flnc_fn,
out_flc_fn, primer_report_fl_fn):
"""
in_read_fn --- a fasta of full-length reads
For each full-length read in in_read_fn FASTA file, detect whether
it is chimeric or not, and write its annotation to
primer_report_fl_fn.
"""
logging.info("Update chimera info to reads annotations " +
"in the output FASTA file and the primer report.")
with FastaReader(in_read_fn) as reader, \
FastaWriter(out_flnc_fn) as writer, \
FastaWriter(out_flc_fn) as writer_chimera, \
open(primer_report_fl_fn, 'w') as reporter:
reporter.write("\t".join(ReadAnnotation.fieldsNames()) + "\n")
for r in reader:
# e.g. r.name="movie/zmw/0_100_CCS fiveend=1;threeend=100;"
readid = r.name.split()[0]
annotation = ReadAnnotation.fromString(
r.name, ignore_polyA=self.ignore_polyA)
if readid not in suspicous_hits: # Non-chimeric reads
# Primer of a primer-trimmed read can not be None.
# assert(annotation.primer is not None)
annotation.chimera = 0
assert (annotation.isFullLength)
self.summary.num_flnc += 1
self.summary.num_flnc_bases += len(r.sequence)
writer.writeRecord(annotation.toAnnotation(), r.sequence)
else: # chimeric reads
annotation.chimera = 1
self.summary.num_flc += 1
writer_chimera.writeRecord(annotation.toAnnotation(),
r.sequence)
reporter.write(annotation.toReportRecord() + "\n")
