def test_reverse(self):
"""Test reverse complement function"""
from mirtop.mirna.realign import reverse_complement
print "Testing ATGC complement"
if "GCAT" != reverse_complement("ATGC"):
logger.error("ATGC complement is not: %s" %
reverse_complement("ATGC"))
def test_reverse(self):
"""Test reverse complement function"""
from mirtop.mirna.realign import reverse_complement
print("Testing ATGC complement")
if "GCAT" != reverse_complement("ATGC"):
raise ValueError("ATGC complement is not: %s" %
reverse_complement("ATGC"))
def _bed(handle, bed_fn):
with open(bed_fn, 'w') as outh:
for line in handle:
if line.startswith("@"):
continue
cols = line.strip().split()
if cols[2]=="*":
logger.debug("READ::Sequence not mapped: %s" % cols[0])
continue
query_name = cols[0]
query_sequence = cols[9]
counts = cols[14]
start = int(cols[3])
strand = "-" if cols[1] == "16" else "+"
chrom = cols[2]
# if there no hits
# if the sequence always matching the read, assuming YES now
# if not current or query_name!=current:
query_sequence = query_sequence if not strand=="-" else reverse_complement(query_sequence)
# logger.debug(("READ::Read name:{0} and Read sequence:{1}").format(line.query_name, sequence))
if query_sequence and query_sequence.find("N") > -1:
continue
end = start + len(query_sequence) - 1
bed_line = "\t".join(list(map(str, [chrom, start, end, query_name,
query_sequence, strand, counts])))
outh.write(bed_line + '\n')
def _analyze_line(line, reads, precursors, handle, args):
if line.reference_id < 0:
logger.debug("READ::Sequence not mapped: %s" % line.reference_id)
return reads
if not line.cigarstring:
logger.debug("READ::Sequence malformed: %s" % line)
return reads
query_name = line.query_name
if query_name not in reads and not line.query_sequence:
return reads
sequence = line.query_sequence if not line.is_reverse else reverse_complement(
line.query_sequence)
logger.debug(("READ::Read name:{0} and Read sequence:{1}").format(
line.query_name, sequence))
if line.query_sequence and line.query_sequence.find("N") > -1:
return reads
if query_name not in reads:
reads[query_name].set_sequence(sequence)
reads[query_name].counts = _get_freq(query_name)
# TODO if args.quant set to 0
# TODO if args.quant increase by 1
if line.is_reverse and not args.genomic:
logger.debug("READ::Sequence is reverse: %s" % line.query_name)
return reads
chrom = handle.getrname(line.reference_id)
start = line.reference_start
cigar = line.cigartuples
# if line.cigarstring.find("I") > -1:
# indels_skip += 1
iso = isomir()
iso.align = line
iso.set_pos(start, len(reads[query_name].sequence))
logger.debug("READ::From BAM start %s end %s at chrom %s" %
(iso.start, iso.end, chrom))
if len(precursors[chrom].replace("N", "")) + 3 < start + len(
reads[query_name].sequence):
logger.debug("READ::%s start + %s sequence size are bigger than"
" size precursor %s" %
(line.reference_id, len(
reads[query_name].sequence), len(precursors[chrom])))
return reads
iso.subs, iso.add, iso.cigar = filter.tune(reads[query_name].sequence,
precursors[chrom], start, cigar)
logger.debug("READ::After tune start %s end %s" % (iso.start, iso.end))
logger.debug("READ::iso add %s iso subs %s" % (iso.add, iso.subs))
reads[query_name].set_precursor(chrom, iso)
return reads
def _analyze_quick_line(line, reads):
if line.reference_id < 0:
logger.debug("READ::Sequence not mapped: %s" % line.reference_id)
return reads
if not line.cigarstring:
logger.debug("READ::Sequence malformed: %s" % line)
return reads
query_name = line.query_name
if query_name not in reads and not line.query_sequence:
return reads
sequence = line.query_sequence if not line.is_reverse else reverse_complement(
line.query_sequence)
logger.debug(("READ::Read name:{0} and Read sequence:{1}").format(
line.query_name, sequence))
if line.query_sequence and line.query_sequence.find("N") > -1:
return reads
if query_name not in reads:
reads[query_name].set_sequence(sequence)
reads[query_name].counts = _get_freq(query_name)
return reads
def _bed(bam_fn, bed_fn):
mode = "r" if bam_fn.endswith("sam") else "rb"
handle = pysam.Samfile(bam_fn, mode)
current = None
if os.path.exists(bed_fn):
return bed_fn
with open(bed_fn, 'w') as outh:
for line in handle:
if line.reference_id < 0:
logger.debug("READ::Sequence not mapped: %s" %
line.reference_id)
continue
if not line.cigarstring:
logger.debug("READ::Sequence malformed: %s" % line)
continue
query_name = line.query_name
if (not current
or query_name != current) and not line.query_sequence:
continue
if not current or query_name != current:
sequence = line.query_sequence if not line.is_reverse else reverse_complement(
line.query_sequence)
logger.debug(("READ::Read name:{0} and Read sequence:{1}").format(
line.query_name, sequence))
if line.query_sequence and line.query_sequence.find("N") > -1:
continue
chrom = handle.getrname(line.reference_id)
start = line.reference_start
end = start + len(sequence) - 1
current = query_name
strand = "+" if not line.is_reverse else "-"
bed_line = "\t".join(
list(
map(str,
[chrom, start, end, query_name, sequence, strand])))
outh.write(bed_line + '\n')