def samblaster_dedup_sort(data, tx_out_file, tx_sr_file, tx_disc_file):
"""Deduplicate and sort with samblaster, produces split read and discordant pair files.
"""
samblaster = config_utils.get_program("samblaster", data["config"])
samtools = config_utils.get_program("samtools", data["config"])
tmp_prefix = "%s-sorttmp" % utils.splitext_plus(tx_out_file)[0]
tobam_cmd = ("{samtools} sort {sort_opt} -@ {cores} -m {mem} -T {tmp_prefix}-{dext} {out_file} -")
# full BAM -- associate more memory and cores
cores, mem = _get_cores_memory(data, downscale=2)
# Potentially downsample to maximum coverage here if not splitting and whole genome sample
ds_cmd = None if data.get("align_split") else bam.get_maxcov_downsample_cl(data, "samtools")
sort_opt = "-n" if data.get("align_split") and dd.get_mark_duplicates(data) else ""
if ds_cmd:
dedup_cmd = "%s %s > %s" % (tobam_cmd.format(out_file="", dext="full", **locals()), ds_cmd, tx_out_file)
else:
dedup_cmd = tobam_cmd.format(out_file="-o %s" % tx_out_file, dext="full", **locals())
# split and discordant BAMs -- give less memory/cores since smaller files
sort_opt = ""
cores, mem = _get_cores_memory(data, downscale=4)
splitter_cmd = tobam_cmd.format(out_file="-o %s" % tx_sr_file, dext="spl", **locals())
discordant_cmd = tobam_cmd.format(out_file="-o %s" % tx_disc_file, dext="disc", **locals())
# samblaster 0.1.22 and better require the -M flag for compatibility with bwa-mem
cmd = ("{samblaster} --addMateTags -M --splitterFile >({splitter_cmd}) --discordantFile >({discordant_cmd}) "
"| {dedup_cmd}")
return cmd.format(**locals())
def samblaster_dedup_sort(data, tx_out_file, tx_sr_file, tx_disc_file):
"""Deduplicate and sort with samblaster, produces split read and discordant pair files.
"""
samblaster = config_utils.get_program("samblaster", data["config"])
samtools = config_utils.get_program("samtools", data["config"])
tmp_prefix = "%s-sorttmp" % utils.splitext_plus(tx_out_file)[0]
tobam_cmd = ("{samtools} sort {sort_opt} -@ {cores} -m {mem} -T {tmp_prefix}-{dext} {out_file} -")
# full BAM -- associate more memory and cores
cores, mem = _get_cores_memory(data, downscale=2)
# Potentially downsample to maximum coverage here if not splitting and whole genome sample
ds_cmd = None if data.get("align_split") else bam.get_maxcov_downsample_cl(data, "samtools")
sort_opt = "-n" if data.get("align_split") and dd.get_mark_duplicates(data) else ""
if ds_cmd:
dedup_cmd = "%s %s > %s" % (tobam_cmd.format(out_file="", dext="full", **locals()), ds_cmd, tx_out_file)
else:
dedup_cmd = tobam_cmd.format(out_file="-o %s" % tx_out_file, dext="full", **locals())
# split and discordant BAMs -- give less memory/cores since smaller files
sort_opt = ""
cores, mem = _get_cores_memory(data, downscale=4)
splitter_cmd = tobam_cmd.format(out_file="-o %s" % tx_sr_file, dext="spl", **locals())
discordant_cmd = tobam_cmd.format(out_file="-o %s" % tx_disc_file, dext="disc", **locals())
# samblaster 0.1.22 and better require the -M flag for compatibility with bwa-mem
cmd = ("{samblaster} --addMateTags -M --splitterFile >({splitter_cmd}) --discordantFile >({discordant_cmd}) "
"| {dedup_cmd}")
return cmd.format(**locals())
def _biobambam_merge_maxcov(data):
"""Combine query sorted BAM files, sort and truncate to maximum coverage.
No de-duplication.
"""
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
return ("bammerge IL={tx_bam_file_list} tmpfile={tx_out_file}-merge %s > {tx_out_file}" % ds_cmd)
def _biobambam_merge_maxcov(data):
"""Combine query sorted BAM files, sort and truncate to maximum coverage.
No de-duplication.
"""
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
return (
"bammerge IL={tx_bam_file_list} tmpfile={tx_out_file}-merge %s > {tx_out_file}"
% ds_cmd)
def _biobambam_merge_dedup_maxcov(data):
"""Combine query sorted BAM files, de-duplicate, sort and truncate to maximum coverage.
Handles split files, checking for large scale whole genome coverage where
we want to downsample to a maximum coverage.
"""
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
return ("bamcat level=0 tmpfile={tx_out_file}-bammerge `cat {tx_bam_file_list}` | "
"bamsormadup threads={num_cores} "
"tmpfile={tx_out_file}-bamsormaduptmp %s > {tx_out_file}" % ds_cmd)
def _biobambam_merge_dedup_maxcov(data):
"""Combine query sorted BAM files, de-duplicate, sort and truncate to maximum coverage.
Handles split files, checking for large scale whole genome coverage where
we want to downsample to a maximum coverage.
"""
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
return (
"bamcat level=0 tmpfile={tx_out_file}-bammerge `cat {tx_bam_file_list}` | "
"bamsormadup threads={num_cores} "
"tmpfile={tx_out_file}-bamsormaduptmp %s > {tx_out_file}" % ds_cmd)
def _biobambam_dedup_sort(data, tx_out_file):
"""Perform streaming deduplication and sorting with biobambam's bamsormadup
"""
samtools = config_utils.get_program("samtools", data["config"])
cores, mem = _get_cores_memory(data, downscale=2)
tmp_file = "%s-sorttmp" % utils.splitext_plus(tx_out_file)[0]
if data.get("align_split"):
cmd = "{samtools} sort -n -@ {cores} -m {mem} -O bam -T {tmp_file}-namesort -o {tx_out_file} -"
else:
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
cmd = ("bamsormadup inputformat=sam threads={cores} tmpfile={tmp_file}-markdup "
"SO=coordinate %s > {tx_out_file}" % ds_cmd)
return cmd.format(**locals())
def _biobambam_dedup_sort(data, tx_out_file):
"""Perform streaming deduplication and sorting with biobambam's bamsormadup
"""
samtools = config_utils.get_program("samtools", data["config"])
cores, mem = _get_cores_memory(data, downscale=2)
tmp_file = "%s-sorttmp" % utils.splitext_plus(tx_out_file)[0]
if data.get("align_split"):
sort_opt = "-n" if data.get("align_split") and dd.get_mark_duplicates(data) else ""
cmd = "{samtools} sort %s -@ {cores} -m {mem} -O bam -T {tmp_file}-namesort -o {tx_out_file} -" % sort_opt
else:
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
cmd = ("bamsormadup inputformat=sam threads={cores} tmpfile={tmp_file}-markdup "
"SO=coordinate %s > {tx_out_file}" % ds_cmd)
return cmd.format(**locals())
def _biobambam_dedup_sort(data, tx_out_file):
"""Perform streaming deduplication and sorting with biobambam's bamsormadup
"""
samtools = config_utils.get_program("samtools", data["config"])
cores, mem = _get_cores_memory(data, downscale=2)
tmp_file = "%s-sorttmp" % utils.splitext_plus(tx_out_file)[0]
if data.get("align_split"):
sort_opt = "-n" if data.get("align_split") and _check_dedup(data) else ""
cmd = "{samtools} sort %s -@ {cores} -m {mem} -O bam -T {tmp_file}-namesort -o {tx_out_file} -" % sort_opt
else:
# scale core usage to avoid memory issues with larger WGS samples
cores = max(1, int(math.ceil(cores * 0.75)))
ds_cmd = bam.get_maxcov_downsample_cl(data, "bamsormadup")
bamsormadup = config_utils.get_program("bamsormadup", data)
cmd = ("{bamsormadup} inputformat=sam threads={cores} tmpfile={tmp_file}-markdup "
"SO=coordinate %s > {tx_out_file}" % ds_cmd)
return cmd.format(**locals())
