def concat_bam(in_fns, out_fn):
"""Concat input bam files to an output bam file.
Note that each input bam has ONLY one reference sequence.
"""
# construct sam header
h = concat_bam_header(in_fns)
o = BamWriter(out_fn, header=h)
for index, in_fn in enumerate(in_fns):
s = Samfile(in_fn, 'rb')
for r in s:
r.tid = index # Overwrite tid !!!
o.write(r)
s.close()
o.close()
def concat_bam(in_fns, out_fn):
"""Concat input bam files to an output bam file.
Note that each input bam has ONLY one reference sequence.
"""
# construct sam header
h = concat_bam_header(in_fns)
o = BamWriter(out_fn, header=h)
for index, in_fn in enumerate(in_fns):
s = pysam.Samfile(in_fn, 'rb')
for r in s:
r.tid = index # Overwrite tid !!!
o.write(r)
s.close()
o.close()
def trim_subreads_and_write(reader,
in_seqids,
out_file,
trim_len,
min_len,
ignore_keyerror=False,
bam=False):
""" Extract (dump) raw subreads of every zmws from in_seqeids from reader
to out_file.
reader --- provides random access to raw subreads in input file.
type = MetaSubreadFastaReader, when input files are in FASTA,
and reads are in format <movie>/<holeNumber>/<subread or CCS>.
type = BamCollection, when input files are in BAM.
trim_len --- trim the first and last n bases when input is BAM
min_len --- minimum read length to write a subread when input is BAM
in_seqids --- zmw ids to dump
out_file --- a FASTA file when input files are in FASTA; a BAM file when
input files are in BAM.
return movies seen
"""
movies = set()
zmw_seen = set()
f = None # output open file handler
if bam:
assert isinstance(reader, BamCollection)
f = BamWriter(out_file, reader.header)
else:
assert isinstance(reader, MetaSubreadFastaReader)
f = FastaWriter(out_file)
for seqid in in_seqids:
zmw = seqid
try:
zmw = '/'.join(seqid.split('/')[0:2])
except ValueError:
raise ValueError("%s does not contain a valid pacbio zmw id." %
seqid)
if zmw not in zmw_seen:
movies.add(zmw.split('/')[0])
zmw_seen.add(zmw)
try:
if bam:
for rec in reader[zmw].subreads:
if len(rec) >= 2 * trim_len + min_len:
f.write(
rec.Clip(rec.readStart + trim_len,
rec.readEnd - trim_len))
else:
for rec in reader[zmw]:
if len(rec) >= 2 * trim_len + min_len:
try:
m, hn, s_e = rec.name.split('/')
s, e = [int(x) for x in s_e.split('_')]
new_id = "%s/%s/%d_%d" % (m, hn, s + trim_len,
e - trim_len)
f.writeRecord(new_id,
rec.sequence[trim_len:-trim_len])
except ValueError:
raise ValueError(
"%s is not a valid pacbio subread." %
rec.name)
except KeyError:
if ignore_keyerror:
logging.warning(
"Ignoring {zmw} because the input FASTA/BAM ".format(
zmw=zmw) + " does not contain it.")
else:
raise ValueError("{0} doesn't exist. Abort!".format(zmw))
f.close()
return movies
def trim_subreads_and_write(reader, in_seqids, out_file, trim_len, min_len,
ignore_keyerror=False, bam=False):
""" Extract (dump) raw subreads of every zmws from in_seqeids from reader
to out_file.
reader --- provides random access to raw subreads in input file.
type = MetaSubreadFastaReader, when input files are in FASTA,
and reads are in format <movie>/<holeNumber>/<subread or CCS>.
type = BamCollection, when input files are in BAM.
trim_len --- trim the first and last n bases when input is BAM
min_len --- minimum read length to write a subread when input is BAM
in_seqids --- zmw ids to dump
out_file --- a FASTA file when input files are in FASTA; a BAM file when
input files are in BAM.
return movies seen
"""
movies = set()
zmw_seen = set()
f = None # output open file handler
if bam:
assert isinstance(reader, BamCollection)
f = BamWriter(out_file, reader.header)
else:
assert isinstance(reader, MetaSubreadFastaReader)
f = FastaWriter(out_file)
for seqid in in_seqids:
zmw = seqid
try:
zmw = '/'.join(seqid.split('/')[0:2])
except ValueError:
raise ValueError("%s does not contain a valid pacbio zmw id." % seqid)
if zmw not in zmw_seen:
movies.add(zmw.split('/')[0])
zmw_seen.add(zmw)
try:
if bam:
for rec in reader[zmw].subreads:
if len(rec) >= 2*trim_len + min_len:
f.write(rec.Clip(rec.readStart+trim_len,
rec.readEnd-trim_len))
else:
for rec in reader[zmw]:
if len(rec) >= 2*trim_len + min_len:
try:
m, hn, s_e = rec.name.split('/')
s, e = [int(x) for x in s_e.split('_')]
new_id = "%s/%s/%d_%d" % (m, hn, s+trim_len, e-trim_len)
f.writeRecord(new_id, rec.sequence[trim_len:-trim_len])
except ValueError:
raise ValueError("%s is not a valid pacbio subread." % rec.name)
except KeyError:
if ignore_keyerror:
logging.warning("Ignoring {zmw} because the input FASTA/BAM ".
format(zmw=zmw) + " does not contain it.")
else:
raise ValueError("{0} doesn't exist. Abort!".format(zmw))
f.close()
return movies