def pesam2bed(input_file,
chrlen_file,
output_file,
logfile,
errfile,
filter_function=check_good_read):
'''
Converts aligned BAM file into BED file of fragments filtering reads that do not pass the filtering criteria
Input:
input_file - aligned (unsorted) paired end BAM
chrlen_file - TSV file of format chromosome<TAB>length
output_file - output file (BED file with start and end corresponding to both ends of the PE fragment)
logfile, errfile - logging files
filter_function - function that accepts an aligned read (pysam.AlignedSegment) and returns true, false or weight of the fragment
Default: utils.check_good_reads
Note: the function can return weight, if trying to accommodate reads from repetitive sequences
Output:
None (written into output file)
See also:
check_good_read
'''
samfile = AlignmentFile(input_file)
take_read = [x for x in itertools.imap(filter_function, samfile)]
# check good_read also checks if this is R1...
samfile.reset()
logfile.write("%d out of %d reads passed filtering\n" %
(sum(take_read), len(take_read) / 2))
logfile.flush()
chrlens = genomic_utils.get_chr_lens(chrlen_file)
outfile = open(output_file, 'w')
count = 0
for i, r in enumerate(samfile):
count += 1
if count % 1000000 == 0:
logfile.write("pesam2bed: %s, Read # %d\n" % (input_file, count))
logfile.flush()
read_1 = r
if take_read[i] == 0:
continue
if not read_1.is_reverse:
rstart = read_1.reference_start
start = max(0, rstart)
rend = rstart + read_1.template_length
end = min(rend, chrlens[samfile.getrname(read_1.reference_id)])
if start > end:
continue
try:
outfile.write(samfile.getrname(read_1.reference_id))
outfile.write('\t')
outfile.write('%d\t%d\t%s\t%d\t+\n' %
(start, end, read_1.qname, take_read[i] * 100))
except ValueError:
errfile.write(
"***** pesam2bed failed with the following value\n")
errfile.write(samfile.getrname(read_1.reference_id))
errfile.write(' %d\t%d\t%s\t%d\t-\n' %
(start, end, read_1.qname, take_read[i] * 100))
errfile.flush()
sys.exit(-1)
else:
rstart = read_1.reference_end - abs(read_1.template_length)
start = max(0, rstart)
rend = read_1.reference_end
end = min(rend, chrlens[samfile.getrname(read_1.reference_id)])
if start > end:
continue
try:
outfile.write(samfile.getrname(read_1.reference_id))
outfile.write('\t')
outfile.write('%d\t%d\t%s\t%d\t-\n' %
(start, end, read_1.qname, take_read[i] * 100))
except ValueError:
errfile.write(
"***** pesam2bed failed with the following value\n")
errfile.write(samfile.getrname(read_1.reference_id))
errfile.write(' %d\t%d\t%s\t%d\t-\n' %
(start, end, read_1.qname, take_read[i] * 100))
errfile.flush()
sys.exit(-1)
outfile.close()
