def _get_bwa_mem_cmd(data,
out_file,
ref_file,
fastq1,
fastq2="",
with_hla=False):
"""Perform piped bwa mem mapping potentially with alternative alleles in GRCh38 + HLA typing.
Commands for HLA post-processing:
base=TEST
run-HLA $base.hla > $base.hla.top
cat $base.hla.HLA*.gt | grep ^GT | cut -f2- > $base.hla.all
rm -f $base.hla.HLA*gt
rm -f $base.hla.HLA*gz
"""
alt_file = ref_file + ".alt"
if with_hla:
bwakit_dir = os.path.dirname(
os.path.realpath(utils.which("run-bwamem")))
hla_base = os.path.join(
utils.safe_makedir(os.path.join(os.path.dirname(out_file), "hla")),
os.path.basename(out_file) + ".hla")
alt_cmd = (
" | {bwakit_dir}/k8 {bwakit_dir}/bwa-postalt.js -p {hla_base} {alt_file}"
)
else:
alt_cmd = ""
if dd.get_aligner(data) == "sentieon-bwa":
bwa_exe = "sentieon-bwa"
exports = sentieon.license_export(data)
else:
bwa_exe = "bwa"
exports = ""
bwa = config_utils.get_program(bwa_exe, data["config"])
num_cores = data["config"]["algorithm"].get("num_cores", 1)
bwa_resources = config_utils.get_resources("bwa", data["config"])
bwa_params = (" ".join([str(x) for x in bwa_resources.get("options", [])])
if "options" in bwa_resources else "")
rg_info = novoalign.get_rg_info(data["rgnames"])
# For UMI runs, pass along consensus tags
c_tags = "-C" if "umi_bam" in data else ""
pairing = "-p" if not fastq2 else ""
# Restrict seed occurances to 1/2 of default, manage memory usage for centromere repeats in hg38
# https://sourceforge.net/p/bio-bwa/mailman/message/31514937/
# http://ehc.ac/p/bio-bwa/mailman/message/32268544/
mem_usage = "-c 250"
bwa_cmd = (
"{exports}{bwa} mem {pairing} {c_tags} {mem_usage} -M -t {num_cores} {bwa_params} -R '{rg_info}' "
"-v 1 {ref_file} {fastq1} {fastq2} ")
return (bwa_cmd + alt_cmd).format(**locals())
def _get_bwa_mem_cmd(data, out_file, ref_file, fastq1, fastq2=""):
"""Perform piped bwa mem mapping potentially with alternative alleles in GRCh38 + HLA typing.
Commands for HLA post-processing:
base=TEST
run-HLA $base.hla > $base.hla.top
cat $base.hla.HLA*.gt | grep ^GT | cut -f2- > $base.hla.all
rm -f $base.hla.HLA*gt
rm -f $base.hla.HLA*gz
"""
alt_file = ref_file + ".alt"
if utils.file_exists(alt_file) and dd.get_hlacaller(data):
bwakit_dir = os.path.dirname(os.path.realpath(utils.which("run-bwamem")))
hla_base = os.path.join(utils.safe_makedir(os.path.join(os.path.dirname(out_file), "hla")),
os.path.basename(out_file) + ".hla")
alt_cmd = (" | {bwakit_dir}/k8 {bwakit_dir}/bwa-postalt.js -p {hla_base} {alt_file}")
else:
alt_cmd = ""
if dd.get_aligner(data) == "sentieon-bwa":
bwa_exe = "sentieon-bwa"
exports = sentieon.license_export(data)
else:
bwa_exe = "bwa"
exports = ""
bwa = config_utils.get_program(bwa_exe, data["config"])
num_cores = data["config"]["algorithm"].get("num_cores", 1)
bwa_resources = config_utils.get_resources("bwa", data["config"])
bwa_params = (" ".join([str(x) for x in bwa_resources.get("options", [])])
if "options" in bwa_resources else "")
rg_info = novoalign.get_rg_info(data["rgnames"])
# For UMI runs, pass along consensus tags
c_tags = "-C" if "umi_bam" in data else ""
pairing = "-p" if not fastq2 else ""
# Restrict seed occurances to 1/2 of default, manage memory usage for centromere repeats in hg38
# https://sourceforge.net/p/bio-bwa/mailman/message/31514937/
# http://ehc.ac/p/bio-bwa/mailman/message/32268544/
mem_usage = "-c 250"
bwa_cmd = ("{exports}{bwa} mem {pairing} {c_tags} {mem_usage} -M -t {num_cores} {bwa_params} -R '{rg_info}' "
"-v 1 {ref_file} {fastq1} {fastq2} ")
return (bwa_cmd + alt_cmd).format(**locals())