def main():
# read params
args = parse_arguments()
log.info('Initializing and making output directory...')
mkdir_p(args.out_dir)
# declare temp arrays
temp_files = [] # files to deleted later at the end
# if bwa index is tarball then unpack it
if args.bwa_index_prefix_or_tar.endswith('.tar') or \
args.bwa_index_prefix_or_tar.endswith('.tar.gz'):
log.info('Unpacking bwa index tar...')
tar = args.bwa_index_prefix_or_tar
# untar
untar(tar, args.out_dir)
bwa_index_prefix = find_bwa_index_prefix(args.out_dir)
temp_files.append('{}*'.format(
bwa_index_prefix))
else:
bwa_index_prefix = args.bwa_index_prefix_or_tar
# check if bowties indices are unpacked on out_dir
chk_bwa_index(bwa_index_prefix)
# bwa
log.info('Running bwa...')
if args.paired_end:
bam = bwa_pe(
args.fastqs[0], args.fastqs[1],
bwa_index_prefix, args.nth, args.mem_gb, args.use_bwa_mem_for_pe,
args.bwa_mem_read_len_limit, args.rescue_reads_for_bwa_mem,
args.out_dir)
else:
bam = bwa_se(
args.fastqs[0],
bwa_index_prefix, args.nth, args.mem_gb,
args.out_dir)
log.info('Removing temporary files...')
rm_f(temp_files)
log.info('Checking if BAM file is empty...')
if bam_is_empty(bam, args.nth):
raise ValueError('BAM file is empty, no reads found.')
log.info('List all files in output directory...')
ls_l(args.out_dir)
log.info('All done.')
def main():
# filt_bam - dupmark_bam - nodup_bam
# \ dup_qc \ pbc_qc
# read params
args = parse_arguments()
log.info('Initializing and making output directory...')
mkdir_p(args.out_dir)
# declare temp arrays
temp_files = [] # files to deleted later at the end
log.info('Removing unmapped/low-quality reads...')
if args.paired_end:
filt_bam = rm_unmapped_lowq_reads_pe(args.bam, args.multimapping,
args.mapq_thresh, args.nth,
args.mem_gb, args.out_dir)
else:
filt_bam = rm_unmapped_lowq_reads_se(args.bam, args.multimapping,
args.mapq_thresh, args.nth,
args.mem_gb, args.out_dir)
log.info('Checking if filtered BAM file is empty...')
if bam_is_empty(filt_bam, args.nth):
help_msg = (
'No reads found in filtered BAM. '
'Low quality sample? '
'Or no reads passing criteria "samtools view -F 1804"? '
'Check samtools flags at '
'https://broadinstitute.github.io/picard/explain-flags.html. ')
if args.paired_end:
help_msg += (
'Or is this truely PE BAM? '
'All unpaired SE reads could be removed by "samtools view -f 2". '
)
raise ValueError(help_msg)
log.info('Marking dupes with {}...'.format(args.dup_marker))
if args.dup_marker == 'picard':
dupmark_bam, dup_qc = mark_dup_picard(filt_bam, args.out_dir,
args.picard_java_heap)
elif args.dup_marker == 'sambamba':
dupmark_bam, dup_qc = mark_dup_sambamba(filt_bam, args.nth,
args.out_dir)
else:
raise argparse.ArgumentTypeError('Unsupported --dup-marker {}'.format(
args.dup_marker))
if args.no_dup_removal:
nodup_bam = filt_bam
else:
temp_files.append(filt_bam)
log.info('Removing dupes...')
if args.paired_end:
nodup_bam = rm_dup_pe(dupmark_bam, args.nth, args.out_dir)
else:
nodup_bam = rm_dup_se(dupmark_bam, args.nth, args.out_dir)
samtools_index(dupmark_bam)
temp_files.append(dupmark_bam + '.bai')
temp_files.append(dupmark_bam)
if len(args.filter_chrs) > 0:
final_bam = remove_chrs_from_bam(nodup_bam, args.filter_chrs,
args.chrsz, args.nth, args.out_dir)
temp_files.append(nodup_bam)
else:
final_bam = nodup_bam
log.info('Checking if final BAM file is empty...')
if bam_is_empty(final_bam, args.nth):
raise ValueError('No reads found in final (filtered/deduped) BAM. '
'Low quality sample? '
'Or BAM with duplicates only? ')
log.info('samtools index (final_bam)...')
samtools_index(final_bam, args.nth, args.out_dir)
log.info('samstat...')
samstat(final_bam, args.nth, args.mem_gb, args.out_dir)
log.info('Generating PBC QC log...')
if args.paired_end:
pbc_qc_pe(dupmark_bam, args.mito_chr_name, args.nth, args.out_dir)
else:
pbc_qc_se(dupmark_bam, args.mito_chr_name, args.out_dir)
log.info('samtools index (raw bam)...')
bam = copy_f_to_dir(args.bam, args.out_dir)
bai = samtools_index(bam, args.nth, args.out_dir)
temp_files.extend([bam, bai])
log.info('Removing temporary files...')
rm_f(temp_files)
log.info('List all files in output directory...')
ls_l(args.out_dir)
log.info('All done.')
def rm_unmapped_lowq_reads_pe(bam, multimapping, mapq_thresh, nth, mem_gb,
out_dir):
"""There are pipes with multiple samtools commands.
For such pipes, use multiple threads (-@) for only one of them.
Priority is on sort > index > fixmate > view.
"""
prefix = os.path.join(out_dir, os.path.basename(strip_ext_bam(bam)))
filt_bam = '{}.filt.bam'.format(prefix)
tmp_filt_bam = '{}.tmp_filt.bam'.format(prefix)
fixmate_bam = '{}.fixmate.bam'.format(prefix)
if multimapping:
run_shell_cmd(
'samtools view -F 524 -f 2 -u {bam} | '
'samtools sort -n /dev/stdin -o {tmp_filt_bam} -T {prefix} {res_param} '
.format(
bam=bam,
tmp_filt_bam=tmp_filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('view', nth=nth),
))
run_shell_cmd(
'samtools view -h {tmp_filt_bam} | '
'$(which assign_multimappers.py) -k {multimapping} --paired-end | '
'samtools fixmate -r /dev/stdin {fixmate_bam} {res_param}'.format(
tmp_filt_bam=tmp_filt_bam,
multimapping=multimapping,
fixmate_bam=fixmate_bam,
res_param=get_samtools_res_param('fixmate', nth=nth),
))
else:
run_shell_cmd(
'samtools view -F 1804 -f 2 -q {mapq_thresh} -u {bam} | '
'samtools sort -n /dev/stdin -o {tmp_filt_bam} -T {prefix} {res_param}'
.format(
mapq_thresh=mapq_thresh,
bam=bam,
tmp_filt_bam=tmp_filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort',
nth=nth,
mem_gb=mem_gb),
))
run_shell_cmd(
'samtools fixmate -r {tmp_filt_bam} {fixmate_bam} {res_param}'.
format(
tmp_filt_bam=tmp_filt_bam,
fixmate_bam=fixmate_bam,
res_param=get_samtools_res_param('fixmate', nth=nth),
))
rm_f(tmp_filt_bam)
run_shell_cmd(
'samtools view -F 1804 -f 2 -u {fixmate_bam} | '
'samtools sort /dev/stdin -o {filt_bam} -T {prefix} {res_param}'.
format(
fixmate_bam=fixmate_bam,
filt_bam=filt_bam,
prefix=prefix,
res_param=get_samtools_res_param('sort', nth=nth, mem_gb=mem_gb),
))
rm_f(fixmate_bam)
log.info('Checking if filtered (but not deduped) BAM is empty '
'after filtering with "samtools view -F 1804 -f 2".')
if bam_is_empty(filt_bam, nth):
raise ValueError(
'No reads found aftering filtering "samtools fixmate"d PE BAM with '
'"samtools view -F 1804 -f 2". '
'Reads are not properly paired even though mapping rate is good? ')
return filt_bam