def mutect2_caller(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Call variation with GATK's MuTect2.
This requires the full non open-source version of GATK 3.5+.
"""
if out_file is None:
out_file = "%s-variants.vcf.gz" % utils.splitext_plus(align_bams[0])[0]
if not utils.file_exists(out_file):
_prep_inputs(align_bams, ref_file, items)
with file_transaction(items[0], out_file) as tx_out_file:
params = ["-T", "MuTect2",
"-R", ref_file,
"--annotation", "ClippingRankSumTest",
"--annotation", "DepthPerSampleHC"]
for a in annotation.get_gatk_annotations(items[0]["config"]):
params += ["--annotation", a]
paired = vcfutils.get_paired_bams(align_bams, items)
params += _add_tumor_params(paired)
params += _add_region_params(region, out_file, items)
params += _add_assoc_params(assoc_files)
params += ["-ploidy", str(ploidy.get_ploidy(items, region))]
resources = config_utils.get_resources("mutect2", items[0]["config"])
if "options" in resources:
params += [str(x) for x in resources.get("options", [])]
broad_runner = broad.runner_from_config(items[0]["config"])
assert LooseVersion(broad_runner.gatk_major_version()) >= LooseVersion("3.5"), \
"Require full version of GATK 3.5+ for mutect2 calling"
broad_runner.new_resources("mutect2")
gatk_cmd = " ".join(broad_runner.cl_gatk(params, os.path.dirname(tx_out_file)))
pp_cmd = _post_process_cl(paired)
cmd = "{gatk_cmd} | {pp_cmd} | bgzip -c > {tx_out_file}"
do.run(cmd.format(**locals()), "MuTect2")
out_file = vcfutils.bgzip_and_index(out_file, items[0]["config"])
return out_file
def _mutect_call_prep(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Preparation work for MuTect.
"""
base_config = items[0]["config"]
broad_runner = broad.runner_from_config(base_config, "mutect")
_check_mutect_version(broad_runner)
broad_runner.run_fn("picard_index_ref", ref_file)
for x in align_bams:
bam.index(x, base_config)
paired = vcfutils.get_paired_bams(align_bams, items)
if not paired:
raise ValueError("Specified MuTect calling but 'tumor' phenotype not present in batch\n"
"https://bcbio-nextgen.readthedocs.org/en/latest/contents/"
"pipelines.html#cancer-variant-calling\n"
"for samples: %s" % ", " .join([dd.get_sample_name(x) for x in items]))
params = ["-R", ref_file, "-T", "MuTect", "-U", "ALLOW_N_CIGAR_READS"]
params += ["--read_filter", "NotPrimaryAlignment"]
params += ["-I:tumor", paired.tumor_bam]
params += ["--tumor_sample_name", paired.tumor_name]
if paired.normal_bam is not None:
params += ["-I:normal", paired.normal_bam]
params += ["--normal_sample_name", paired.normal_name]
if paired.normal_panel is not None:
params += ["--normal_panel", paired.normal_panel]
params += _config_params(base_config, assoc_files, region, out_file)
return broad_runner, params
def run(items, background=None):
"""Detect copy number variations from batched set of samples using CNVkit.
"""
if not background: background = []
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
if paired:
inputs = [paired.tumor_data]
background_bams = [paired.normal_bam]
background_names = [paired.normal_name]
else:
inputs = [items]
background_bams = [x["align_bam"] for x in background]
background_names = [dd.get_sample_name(x) for x in background]
orig_vcf_file = _run_wham(inputs, background_bams)
wclass_vcf_file = _add_wham_classification(orig_vcf_file, inputs)
vcf_file = _fix_vcf(wclass_vcf_file, inputs, background_names)
bed_file = _convert_to_bed(vcf_file, inputs)
out = []
for data in items:
if "sv" not in data:
data["sv"] = []
data["sv"].append({
"variantcaller": "wham",
"vrn_file": vcf_file,
"bed_file": bed_file
})
out.append(data)
return out
def run(items, background=None):
"""Detect copy number variations from batched set of samples using WHAM.
"""
if not background: background = []
background_bams = []
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
if paired:
inputs = [paired.tumor_data]
if paired.normal_bam:
background = [paired.normal_data]
background_bams = [paired.normal_bam]
else:
assert not background
inputs, background = shared.find_case_control(items)
background_bams = [x["align_bam"] for x in background]
orig_vcf = _run_wham(inputs, background_bams)
out = []
for data in inputs:
if "sv" not in data:
data["sv"] = []
sample_vcf = "%s-%s.vcf.gz" % (utils.splitext_plus(orig_vcf)[0], dd.get_sample_name(data))
sample_vcf = vcfutils.select_sample(orig_vcf, dd.get_sample_name(data), sample_vcf, data["config"])
if background:
sample_vcf = filter_by_background(sample_vcf, orig_vcf, background, data)
data["sv"].append({"variantcaller": "wham",
"vrn_file": sample_vcf})
out.append(data)
return out
def _mutect_call_prep(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Preparation work for MuTect.
"""
base_config = items[0]["config"]
broad_runner = broad.runner_from_config(base_config, "mutect")
_check_mutect_version(broad_runner)
broad_runner.run_fn("picard_index_ref", ref_file)
for x in align_bams:
bam.index(x, base_config)
paired = vcfutils.get_paired_bams(align_bams, items)
params = ["-R", ref_file, "-T", "MuTect", "-U", "ALLOW_N_CIGAR_READS"]
# if coverage_depth_max is not given, default to 10000
downsample_cov = get_in(paired.tumor_config,
("algorithm", "coverage_depth_max"), 10000)
# if coverage_depth_max is zero, default to Broad default value (currently 1500)
params += ["--downsample_to_coverage",
max(1500, downsample_cov)] if downsample_cov > 0 else []
params += ["--read_filter", "NotPrimaryAlignment"]
params += ["-I:tumor", paired.tumor_bam]
params += ["--tumor_sample_name", paired.tumor_name]
if paired.normal_bam is not None:
params += ["-I:normal", paired.normal_bam]
params += ["--normal_sample_name", paired.normal_name]
if paired.normal_panel is not None:
params += ["--normal_panel", paired.normal_panel]
params += _config_params(base_config, assoc_files, region, out_file)
return broad_runner, params
def _SID_call_prep(align_bams, items, ref_file, assoc_files, region=None, out_file=None):
"""Preparation work for SomaticIndelDetector.
"""
base_config = items[0]["config"]
for x in align_bams:
bam.index(x, base_config)
params = ["-R", ref_file, "-T", "SomaticIndelDetector", "-U", "ALLOW_N_CIGAR_READS"]
# Limit per base read start count to between 200-10000, i.e. from any base
# can no more 10000 new reads begin.
# Further, limit maxNumberOfReads accordingly, otherwise SID discards
# windows for high coverage panels.
window_size = 200 # default SID value
paired = vcfutils.get_paired_bams(align_bams, items)
max_depth = min(max(200, get_in(paired.tumor_config,
("algorithm", "coverage_depth_max"), 10000)), 10000)
params += ["--downsample_to_coverage", max_depth]
params += ["--maxNumberOfReads", str(int(max_depth) * window_size)]
params += ["--read_filter", "NotPrimaryAlignment"]
params += ["-I:tumor", paired.tumor_bam]
min_af = float(get_in(paired.tumor_config, ("algorithm", "min_allele_fraction"), 10)) / 100.0
if paired.normal_bam is not None:
params += ["-I:normal", paired.normal_bam]
# notice there must be at least 4 reads of coverage in normal
params += ["--filter_expressions", "T_COV<6||N_COV<4||T_INDEL_F<%s||T_INDEL_CF<0.7" % min_af]
else:
params += ["--unpaired"]
params += ["--filter_expressions", "COV<6||INDEL_F<%s||INDEL_CF<0.7" % min_af]
if region:
params += ["-L", bamprep.region_to_gatk(region), "--interval_set_rule",
"INTERSECTION"]
return params
def estimate(items, batch, config):
"""Estimate heterogeneity for a pair of tumor/normal samples. Run in parallel.
"""
hetcallers = {
"theta": theta.run,
"phylowgs": phylowgs.run,
"bubbletree": bubbletree.run
}
paired = vcfutils.get_paired_bams([dd.get_align_bam(d) for d in items],
items)
calls = _get_calls(paired.tumor_data)
variants = get_variants(paired.tumor_data)
het_info = []
for hetcaller in _get_hetcallers(items):
try:
hetfn = hetcallers[hetcaller]
except KeyError:
hetfn = None
print("%s not yet implemented" % hetcaller)
if hetfn:
hetout = hetfn(variants[0], calls, paired)
if hetout:
het_info.append(hetout)
out = []
for data in items:
if batch == _get_batches(data)[0]:
if dd.get_sample_name(data) == paired.tumor_name:
if het_info:
data["heterogeneity"] = het_info
out.append([data])
return out
def _run_scalpel_paired(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Detect indels with Scalpel.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(config, out_file) as tx_out_file:
paired = get_paired_bams(align_bams, items)
if not paired.normal_bam:
ann_file = _run_scalpel_caller(align_bams, items, ref_file,
assoc_files, region, out_file)
return ann_file
vcfstreamsort = config_utils.get_program("vcfstreamsort", config)
perl_exports = utils.get_perl_exports(os.path.dirname(tx_out_file))
tmp_path = "%s-scalpel-work" % utils.splitext_plus(out_file)[0]
db_file = os.path.join(tmp_path, "main", "somatic.db")
if not os.path.exists(db_file + ".dir"):
if os.path.exists(tmp_path):
utils.remove_safe(tmp_path)
opts = " ".join(_scalpel_options_from_config(items, config, out_file, region, tmp_path))
opts += " --ref {}".format(ref_file)
opts += " --dir %s" % tmp_path
# caling
cl = ("{perl_exports} && "
"scalpel-discovery --somatic {opts} --tumor {paired.tumor_bam} --normal {paired.normal_bam}")
do.run(cl.format(**locals()), "Genotyping paired variants with Scalpel", {})
# filtering to adjust input parameters
bed_opts = " ".join(_scalpel_bed_file_opts(items, config, out_file, region, tmp_path))
use_defaults = True
if use_defaults:
scalpel_tmp_file = os.path.join(tmp_path, "main/somatic.indel.vcf")
# Uses default filters but can tweak min-alt-count-tumor and min-phred-fisher
# to swap precision for sensitivity
else:
scalpel_tmp_file = os.path.join(tmp_path, "main/somatic-indel-filter.vcf.gz")
with file_transaction(config, scalpel_tmp_file) as tx_indel_file:
cmd = ("{perl_exports} && "
"scalpel-export --somatic {bed_opts} --ref {ref_file} --db {db_file} "
"--min-alt-count-tumor 5 --min-phred-fisher 10 --min-vaf-tumor 0.1 "
"| bgzip -c > {tx_indel_file}")
do.run(cmd.format(**locals()), "Scalpel somatic indel filter", {})
scalpel_tmp_file = bgzip_and_index(scalpel_tmp_file, config)
scalpel_tmp_file_common = bgzip_and_index(os.path.join(tmp_path, "main/common.indel.vcf"), config)
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
bcftools_cmd_chi2 = get_scalpel_bcftools_filter_expression("chi2", config)
bcftools_cmd_common = get_scalpel_bcftools_filter_expression("reject", config)
fix_ambig = vcfutils.fix_ambiguous_cl()
cl2 = ("vcfcat <({bcftools_cmd_chi2} {scalpel_tmp_file}) "
"<({bcftools_cmd_common} {scalpel_tmp_file_common}) | "
" {fix_ambig} | {vcfstreamsort} {compress_cmd} > {tx_out_file}")
do.run(cl2.format(**locals()), "Finalising Scalpel variants", {})
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"), ref_file,
config)
return ann_file
def _paired_load_script(work_bams, names, chrom, pairmode, items):
"""Prepare BAMs for assessing CNVs in a paired tumor/normal setup.
"""
paired = vcfutils.get_paired_bams(work_bams, items)
return _paired_prep.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode)
def _SID_call_prep(align_bams, items, ref_file, assoc_files, region=None, out_file=None):
"""Preparation work for SomaticIndelDetector.
"""
base_config = items[0]["config"]
for x in align_bams:
bam.index(x, base_config)
params = ["-R", ref_file, "-T", "SomaticIndelDetector", "-U", "ALLOW_N_CIGAR_READS"]
# Limit per base read start count to between 200-10000, i.e. from any base
# can no more 10000 new reads begin.
# Further, limit maxNumberOfReads accordingly, otherwise SID discards
# windows for high coverage panels.
window_size = 200 # default SID value
paired = vcfutils.get_paired_bams(align_bams, items)
max_depth = min(max(200, get_in(paired.tumor_config,
("algorithm", "coverage_depth_max"), 10000)), 10000)
params += ["--downsample_to_coverage", max_depth]
params += ["--maxNumberOfReads", str(int(max_depth) * window_size)]
params += ["--read_filter", "NotPrimaryAlignment"]
params += ["-I:tumor", paired.tumor_bam]
min_af = float(get_in(paired.tumor_config, ("algorithm", "min_allele_fraction"), 10)) / 100.0
if paired.normal_bam is not None:
params += ["-I:normal", paired.normal_bam]
# notice there must be at least 4 reads of coverage in normal
params += ["--filter_expressions", "T_COV<6||N_COV<4||T_INDEL_F<%s||T_INDEL_CF<0.7" % min_af]
else:
params += ["--unpaired"]
params += ["--filter_expressions", "COV<6||INDEL_F<%s||INDEL_CF<0.7" % min_af]
if region:
params += ["-L", bamprep.region_to_gatk(region), "--interval_set_rule",
"INTERSECTION"]
return params
def run_tnhaplotyper(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Call variants with Sentieon's TNhaplotyper (MuTect2 like).
"""
if out_file is None:
out_file = "%s-variants.vcf.gz" % utils.splitext_plus(align_bams[0])[0]
if not utils.file_exists(out_file):
variant_regions = bedutils.merge_overlaps(dd.get_variant_regions(items[0]), items[0])
interval = _get_interval(variant_regions, region, out_file, items)
with file_transaction(items[0], out_file) as tx_out_file:
paired = vcfutils.get_paired_bams(align_bams, items)
assert paired.normal_bam, "Require normal BAM for Sentieon TNhaplotyper"
dbsnp = "--dbsnp %s" % (assoc_files.get("dbsnp")) if "dbsnp" in assoc_files else ""
cosmic = "--cosmic %s" % (assoc_files.get("cosmic")) if "cosmic" in assoc_files else ""
license = license_export(items[0])
tx_orig_file = "%s-orig%s" % utils.splitext_plus(tx_out_file)
cores = dd.get_num_cores(items[0])
cmd = ("{license}sentieon driver -t {cores} -r {ref_file} "
"-i {paired.tumor_bam} -i {paired.normal_bam} {interval} "
"--algo TNhaplotyper "
"--tumor_sample {paired.tumor_name} --normal_sample {paired.normal_name} "
"{dbsnp} {cosmic} {tx_orig_file}")
do.run(cmd.format(**locals()), "Sentieon TNhaplotyper")
cmd = ("gunzip -c {tx_orig_file} | "
"sed 's/ID=ECNT,Number=1,Type=Integer/ID=ECNT,Number=1,Type=String/' | "
"sed 's/ID=HCNT,Number=1,Type=Integer/ID=HCNT,Number=1,Type=String/' | "
"sed 's/ID=NLOD,Number=1,Type=Float/ID=NLOD,Number=1,Type=String/' | "
"sed 's/ID=TLOD,Number=1,Type=Float/ID=TLOD,Number=1,Type=String/' | "
"sed 's/ID=PON,Number=1,Type=Integer/ID=PON,Number=1,Type=String/' | "
"bgzip -c > {tx_out_file}")
do.run(cmd.format(**locals()), "Sentieon TNhaplotyper: make headers GATK compatible")
vcfutils.bgzip_and_index(tx_out_file, items[0]["config"])
return out_file
def estimate(items, batch, config):
"""Estimate heterogeneity for a pair of tumor/normal samples. Run in parallel.
"""
hetcallers = {"theta": theta.run,
"phylowgs": phylowgs.run,
"bubbletree": bubbletree.run}
paired = vcfutils.get_paired_bams([dd.get_align_bam(d) for d in items], items)
calls = _get_calls(paired.tumor_data)
variants = _get_variants(paired.tumor_data)
het_info = []
for hetcaller in _get_hetcallers(items):
try:
hetfn = hetcallers[hetcaller]
except KeyError:
hetfn = None
print "%s not yet implemented" % hetcaller
if hetfn:
hetout = hetfn(variants[0], calls, paired)
if hetout:
het_info.append(hetout)
out = []
for data in items:
if batch == _get_batches(data)[0]:
if dd.get_sample_name(data) == paired.tumor_name:
if het_info:
data["heterogeneity"] = het_info
out.append([data])
return out
def run(items):
"""Perform detection of structural variations with lumpy, using bwa-mem alignment.
"""
if not all(utils.get_in(data, ("config", "algorithm", "aligner")) in ["bwa", False, None] for data in items):
raise ValueError("Require bwa-mem alignment input for lumpy structural variation detection")
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data if paired and paired.tumor_data else items[0])
full_bams, sr_bams, disc_bams = [], [], []
for data in items:
dedup_bam, sr_bam, disc_bam = sshared.get_split_discordants(data, work_dir)
full_bams.append(dedup_bam)
sr_bams.append(sr_bam)
disc_bams.append(disc_bam)
lumpy_vcf, exclude_file = _run_lumpy(full_bams, sr_bams, disc_bams, work_dir, items)
out = []
for i, data in enumerate(items):
if "sv" not in data:
data["sv"] = []
sample = dd.get_sample_name(data)
dedup_bam, sr_bam, _ = sshared.get_split_discordants(data, work_dir)
sample_vcf = vcfutils.select_sample(lumpy_vcf, sample,
utils.append_stem(lumpy_vcf, "-%s" % sample),
data["config"])
gt_vcf = _run_svtyper(sample_vcf, dedup_bam, sr_bam, data)
filter_vcf = _filter_by_support(gt_vcf, data)
data["sv"].append({"variantcaller": "lumpy",
"vrn_file": filter_vcf,
"exclude_file": exclude_file})
out.append(data)
return out
def _run_freebayes_paired(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Detect SNPs and indels with FreeBayes.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(out_file) as tx_out_file:
paired = get_paired_bams(align_bams, items)
if not paired.normal_bam:
raise ValueError("Require both tumor and normal BAM files for FreeBayes cancer calling")
vcfsamplediff = config_utils.get_program("vcfsamplediff", config)
freebayes = config_utils.get_program("freebayes", config)
opts = " ".join(_freebayes_options_from_config(items, config, out_file, region))
opts += " -f {}".format(ref_file)
# NOTE: The first sample name in the vcfsamplediff call is
# the one supposed to be the *germline* one
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
cl = ("{freebayes} --pooled-discrete --pvar 0.7"
" --genotype-qualities {opts} {paired.tumor_bam}"
" {paired.normal_bam} | {vcfsamplediff} -s VT"
" {paired.normal_name} {paired.tumor_name}"
" - {compress_cmd} > {tx_out_file}")
bam.index(paired.tumor_bam, config)
bam.index(paired.normal_bam, config)
do.run(cl.format(**locals()), "Genotyping paired variants with FreeBayes", {})
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files["dbsnp"], ref_file,
config)
return ann_file
def run(items, background=None):
"""Detect copy number variations from batched set of samples using WHAM.
"""
if not background: background = []
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
if paired:
inputs = [paired.tumor_data]
background_bams = [paired.normal_bam]
background_names = [paired.normal_name]
else:
assert not background
inputs, background = shared.find_case_control(items)
background_bams = [x["align_bam"] for x in background]
background_names = [dd.get_sample_name(x) for x in background]
orig_vcf_file = _run_wham(inputs, background_bams)
wclass_vcf_file = _add_wham_classification(orig_vcf_file, inputs)
vcf_file = _fix_vcf(wclass_vcf_file, inputs, background_names)
bed_file = _convert_to_bed(vcf_file, inputs, use_lrt=len(background_bams) > 0)
out = []
for data in items:
if "sv" not in data:
data["sv"] = []
data["sv"].append({"variantcaller": "wham",
"vrn_file": _subset_to_sample(bed_file, data),
"vcf_file": vcf_file})
out.append(data)
return out
def _run_qsnp_paired(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Detect somatic mutations with qSNP.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf" % os.path.splitext(align_bams[0])[0]
if not utils.file_exists(out_file):
out_file = out_file.replace(".gz", "")
with file_transaction(config, out_file) as tx_out_file:
with tx_tmpdir(config) as tmpdir:
with utils.chdir(tmpdir):
paired = get_paired_bams(align_bams, items)
qsnp = config_utils.get_program("qsnp", config)
resources = config_utils.get_resources("qsnp", config)
mem = " ".join(resources.get("jvm_opts", ["-Xms750m -Xmx4g"]))
qsnp_log = os.path.join(tmpdir, "qsnp.log")
qsnp_init = os.path.join(tmpdir, "qsnp.ini")
if region:
paired = _create_bam_region(paired, region, tmpdir)
_create_input(paired, tx_out_file, ref_file, assoc_files['dbsnp'], qsnp_init)
cl = ("{qsnp} {mem} -i {qsnp_init} -log {qsnp_log}")
do.run(cl.format(**locals()), "Genotyping paired variants with Qsnp", {})
out_file = _filter_vcf(out_file)
out_file = bgzip_and_index(out_file, config)
return out_file
def run_qsnp(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Run qSNP calling on paired tumor/normal.
"""
if utils.file_exists(out_file):
return out_file
paired = get_paired_bams(align_bams, items)
if paired.normal_bam:
region_files = []
regions = _clean_regions(items, region)
if regions:
for region in regions:
out_region_file = out_file.replace(".vcf.gz",
_to_str(region) + ".vcf.gz")
region_file = _run_qsnp_paired(align_bams, items, ref_file,
assoc_files, region,
out_region_file)
region_files.append(region_file)
out_file = combine_variant_files(region_files, out_file, ref_file,
items[0]["config"])
if not region:
out_file = _run_qsnp_paired(align_bams, items, ref_file,
assoc_files, region, out_file)
return out_file
else:
raise ValueError("qSNP only works on paired samples")
def run_freebayes(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Run FreeBayes variant calling, either paired tumor/normal or germline calling.
"""
if is_paired_analysis(align_bams, items):
paired = get_paired_bams(align_bams, items)
if not paired.normal_bam:
call_file = _run_freebayes_caller(align_bams,
items,
ref_file,
assoc_files,
region,
out_file,
somatic=paired)
else:
call_file = _run_freebayes_paired(align_bams, items, ref_file,
assoc_files, region, out_file)
else:
vcfutils.check_paired_problems(items)
call_file = _run_freebayes_caller(align_bams, items, ref_file,
assoc_files, region, out_file)
return call_file
def run(items, background=None):
"""Detect copy number variations from batched set of samples using WHAM.
"""
if not background: background = []
background_bams = []
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
if paired:
inputs = [paired.tumor_data]
if paired.normal_bam:
background = [paired.normal_data]
background_bams = [paired.normal_bam]
else:
assert not background
inputs, background = shared.find_case_control(items)
background_bams = [x["align_bam"] for x in background]
orig_vcf = _run_wham(inputs, background_bams)
out = []
for data in inputs:
if "sv" not in data:
data["sv"] = []
sample_vcf = "%s-%s.vcf.gz" % (utils.splitext_plus(orig_vcf)[0],
dd.get_sample_name(data))
sample_vcf = vcfutils.select_sample(orig_vcf, dd.get_sample_name(data),
sample_vcf, data["config"])
effects_vcf, _ = effects.add_to_vcf(sample_vcf, data, "snpeff")
data["sv"].append({
"variantcaller": "wham",
"vrn_file": effects_vcf or sample_vcf
})
out.append(data)
return out
def run(items, background=None):
"""Detect copy number variations from tumor/normal samples using Battenberg.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
if not paired or not paired.normal_bam:
logger.warn(
"Battenberg only works on paired tumor/normal inputs, skipping %s"
% dd.get_sample_name(items[0]))
batout = None
elif not tz.get_in(["genome_resources", "aliases", "human"],
paired.tumor_data):
logger.warn("Battenberg only works on human data, skipping %s" %
dd.get_sample_name(items[0]))
batout = None
else:
batout = _do_run(paired)
batout["variantcaller"] = "battenberg"
out = []
for data in items:
if batout:
if "sv" not in data:
data["sv"] = []
data["sv"].append(batout)
out.append(data)
return out
def _mutect_call_prep(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Preparation work for MuTect.
"""
base_config = items[0]["config"]
broad_runner = broad.runner_from_config(base_config, "mutect")
_check_mutect_version(broad_runner)
broad_runner.run_fn("picard_index_ref", ref_file)
for x in align_bams:
bam.index(x, base_config)
paired = vcfutils.get_paired_bams(align_bams, items)
params = ["-R", ref_file, "-T", "MuTect", "-U", "ALLOW_N_CIGAR_READS"]
params += ["--downsample_to_coverage", max(200, get_in(paired.tumor_config,
("algorithm", "coverage_depth_max"), 10000))]
params += ["--read_filter", "NotPrimaryAlignment"]
params += ["-I:tumor", paired.tumor_bam]
params += ["--tumor_sample_name", paired.tumor_name]
if paired.normal_bam is not None:
params += ["-I:normal", paired.normal_bam]
params += ["--normal_sample_name", paired.normal_name]
if paired.normal_panel is not None:
params += ["--normal_panel", paired.normal_panel]
params += _config_params(base_config, assoc_files, region, out_file)
return broad_runner, params
def _run_freebayes_paired(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Detect SNPs and indels with FreeBayes.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(
align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(out_file) as tx_out_file:
paired = get_paired_bams(align_bams, items)
if not paired.normal_bam:
raise ValueError(
"Require both tumor and normal BAM files for FreeBayes cancer calling"
)
vcfsamplediff = config_utils.get_program("vcfsamplediff", config)
vcffilter = config_utils.get_program("vcffilter", config)
freebayes = config_utils.get_program("freebayes", config)
opts = " ".join(
_freebayes_options_from_config(items, config, out_file,
region))
opts += " -f {}".format(ref_file)
if "--min-alternate-fraction" not in opts and "-F" not in opts:
# add minimum reportable allele frequency
# FreeBayes defaults to 20%, but use 10% by default for the
# tumor case
min_af = float(
utils.get_in(paired.tumor_config,
("algorithm", "min_allele_fraction"),
10)) / 100.0
opts += " --min-alternate-fraction %s" % min_af
# NOTE: The first sample name in the vcfsamplediff call is
# the one supposed to be the *germline* one
# NOTE: -s in vcfsamplediff (strict checking: i.e., require no
# reads in the germline to call somatic) is not used as it is
# too stringent
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
cl = (
"{freebayes} --pooled-discrete --genotype-qualities "
"{opts} {paired.tumor_bam} {paired.normal_bam} "
"| {vcffilter} -f 'QUAL > 1' -s "
"| {vcfsamplediff} VT {paired.normal_name} {paired.tumor_name} - "
"{compress_cmd} > {tx_out_file}")
bam.index(paired.tumor_bam, config)
bam.index(paired.normal_bam, config)
do.run(cl.format(**locals()),
"Genotyping paired variants with FreeBayes", {})
fix_somatic_calls(out_file, config)
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files["dbsnp"], ref_file,
config)
return ann_file
def run_tnscope(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Call variants with Sentieon's TNscope somatic caller.
"""
if out_file is None:
out_file = "%s-variants.vcf.gz" % utils.splitext_plus(align_bams[0])[0]
if not utils.file_exists(out_file):
variant_regions = bedutils.merge_overlaps(
dd.get_variant_regions(items[0]), items[0])
interval = _get_interval(variant_regions, region, out_file, items)
with file_transaction(items[0], out_file) as tx_out_file:
paired = vcfutils.get_paired_bams(align_bams, items)
assert paired and paired.normal_bam, "Require normal BAM for Sentieon TNscope"
dbsnp = "--dbsnp %s" % (
assoc_files.get("dbsnp")) if "dbsnp" in assoc_files else ""
license = license_export(items[0])
cores = dd.get_num_cores(items[0])
cmd = (
"{license}sentieon driver -t {cores} -r {ref_file} "
"-i {paired.tumor_bam} -i {paired.normal_bam} {interval} "
"--algo TNscope "
"--tumor_sample {paired.tumor_name} --normal_sample {paired.normal_name} "
"{dbsnp} {tx_out_file}")
do.run(cmd.format(**locals()), "Sentieon TNscope")
return out_file
def _run_freebayes_paired(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Detect SNPs and indels with FreeBayes for paired tumor/normal samples.
Sources of options for FreeBayes:
mailing list: https://groups.google.com/d/msg/freebayes/dTWBtLyM4Vs/HAK_ZhJHguMJ
mailing list: https://groups.google.com/forum/#!msg/freebayes/LLH7ZfZlVNs/63FdD31rrfEJ
speedseq: https://github.com/cc2qe/speedseq/blob/e6729aa2589eca4e3a946f398c1a2bdc15a7300d/bin/speedseq#L916
sga/freebayes: https://github.com/jts/sga-extra/blob/7e28caf71e8107b697f9be7162050e4fa259694b/
sga_generate_varcall_makefile.pl#L299
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(
align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(items[0], out_file) as tx_out_file:
paired = get_paired_bams(align_bams, items)
assert paired.normal_bam, "Require normal BAM for FreeBayes paired calling and filtering"
freebayes = config_utils.get_program("freebayes", config)
opts, no_target_regions = _freebayes_options_from_config(
items, config, out_file, region)
if no_target_regions:
vcfutils.write_empty_vcf(
tx_out_file,
config,
samples=[
x for x in [paired.tumor_name, paired.normal_name] if x
])
else:
opts = " ".join(opts)
opts += " --min-repeat-entropy 1"
opts += " --no-partial-observations"
opts = _add_somatic_opts(opts, paired)
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
fix_ambig = vcfutils.fix_ambiguous_cl()
clean_fmt_cmd = _clean_freebayes_fmt_cl()
py_cl = os.path.join(os.path.dirname(sys.executable), "py")
cl = (
"{freebayes} -f {ref_file} {opts} "
"{paired.tumor_bam} {paired.normal_bam} "
"""| bcftools filter -i 'ALT="<*>" || QUAL > 5' """
"| {py_cl} -x 'bcbio.variation.freebayes.call_somatic(x)' "
"| {fix_ambig} | {clean_fmt_cmd} bcftools view -a - | "
"{py_cl} -x 'bcbio.variation.freebayes.remove_missingalt(x)' | "
"vcfallelicprimitives -t DECOMPOSED --keep-geno | vcffixup - | vcfstreamsort | "
"vt normalize -n -r {ref_file} -q - | vcfuniqalleles "
"{compress_cmd} > {tx_out_file}")
do.run(cl.format(**locals()),
"Genotyping paired variants with FreeBayes", {})
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"),
ref_file, config)
return ann_file
def _ready_for_het_analysis(items):
"""Check if a sample has input information for heterogeneity analysis.
We currently require a tumor/normal sample containing both CNV and variant calls.
"""
paired = vcfutils.get_paired_bams([dd.get_align_bam(d) for d in items], items)
if paired and paired.normal_bam:
return _get_variants(paired.tumor_data) and _get_cnvs(paired.tumor_data)
def mutect2_caller(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Call variation with GATK's MuTect2.
This requires the full non open-source version of GATK 3.5+.
"""
if out_file is None:
out_file = "%s-variants.vcf.gz" % utils.splitext_plus(align_bams[0])[0]
if not utils.file_exists(out_file):
paired = vcfutils.get_paired_bams(align_bams, items)
broad_runner = broad.runner_from_config(items[0]["config"])
gatk_type = broad_runner.gatk_type()
_prep_inputs(align_bams, ref_file, items)
with file_transaction(items[0], out_file) as tx_out_file:
params = [
"-T", "Mutect2" if gatk_type == "gatk4" else "MuTect2", "-R",
ref_file, "--annotation", "ClippingRankSumTest",
"--annotation", "DepthPerSampleHC"
]
for a in annotation.get_gatk_annotations(
items[0]["config"], include_baseqranksum=False):
params += ["--annotation", a]
# Avoid issues with BAM CIGAR reads that GATK doesn't like
if gatk_type == "gatk4":
params += ["--read-validation-stringency", "LENIENT"]
params += _add_tumor_params(paired, items, gatk_type)
params += _add_region_params(region, out_file, items, gatk_type)
# Avoid adding dbSNP/Cosmic so they do not get fed to variant filtering algorithm
# Not yet clear how this helps or hurts in a general case.
#params += _add_assoc_params(assoc_files)
params += ["-ploidy", str(ploidy.get_ploidy(items, region))]
resources = config_utils.get_resources("mutect2",
items[0]["config"])
if "options" in resources:
params += [str(x) for x in resources.get("options", [])]
assert LooseVersion(broad_runner.gatk_major_version()) >= LooseVersion("3.5"), \
"Require full version of GATK 3.5+ for mutect2 calling"
broad_runner.new_resources("mutect2")
gatk_cmd = broad_runner.cl_gatk(params,
os.path.dirname(tx_out_file))
if gatk_type == "gatk4":
tx_raw_prefilt_file = "%s-raw%s" % utils.splitext_plus(
tx_out_file)
tx_raw_file = "%s-raw-filt%s" % utils.splitext_plus(
tx_out_file)
filter_cmd = _mutect2_filter(broad_runner, tx_raw_prefilt_file,
tx_raw_file)
cmd = "{gatk_cmd} -O {tx_raw_prefilt_file} && {filter_cmd}"
else:
tx_raw_file = "%s-raw%s" % utils.splitext_plus(tx_out_file)
cmd = "{gatk_cmd} > {tx_raw_file}"
do.run(cmd.format(**locals()), "MuTect2")
out_file = _af_filter(paired.tumor_data, tx_raw_file, out_file)
return vcfutils.bgzip_and_index(out_file, items[0]["config"])
def _run_cnvkit_cancer(items, background, access_file, work_dir):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
return _run_cnvkit_shared(items[0], [paired.tumor_bam],
[paired.normal_bam],
access_file,
work_dir,
background_name=paired.normal_name)
def run(items):
"""Perform detection of structural variations with lumpy, using bwa-mem alignment.
"""
if not all(utils.get_in(data, ("config", "algorithm", "aligner"))
in ["bwa", "sentieon-bwa", False, None] for data in items):
raise ValueError("Require bwa-mem alignment input for lumpy structural variation detection")
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data if paired and paired.tumor_data else items[0])
previous_evidence = {}
full_bams, sr_bams, disc_bams = [], [], []
for data in items:
sr_bam, disc_bam = sshared.get_split_discordants(data, work_dir)
full_bams.append(dd.get_align_bam(data))
sr_bams.append(sr_bam)
disc_bams.append(disc_bam)
cur_dels, cur_dups = _bedpes_from_cnv_caller(data, work_dir)
previous_evidence[dd.get_sample_name(data)] = {}
if cur_dels and utils.file_exists(cur_dels):
previous_evidence[dd.get_sample_name(data)]["dels"] = cur_dels
if cur_dups and utils.file_exists(cur_dups):
previous_evidence[dd.get_sample_name(data)]["dups"] = cur_dups
lumpy_vcf, exclude_file = _run_lumpy(full_bams, sr_bams, disc_bams, previous_evidence,
work_dir, items)
gt_vcfs = {}
for data in items:
sample = dd.get_sample_name(data)
sample_vcf = vcfutils.select_sample(lumpy_vcf, sample,
utils.append_stem(lumpy_vcf, "-%s" % sample),
data["config"])
if "bnd-genotype" in dd.get_tools_on(data):
gt_vcf = _run_svtyper(sample_vcf, dd.get_align_bam(data), exclude_file, data)
elif "lumpy-genotype" in dd.get_tools_off(data):
gt_vcf = sample_vcf
else:
std_vcf, bnd_vcf = _split_breakends(sample_vcf, data)
std_gt_vcf = _run_svtyper(std_vcf, dd.get_align_bam(data), exclude_file, data)
gt_vcf = vcfutils.concat_variant_files_bcftools(
orig_files=[std_gt_vcf, bnd_vcf],
out_file="%s-combined.vcf.gz" % utils.splitext_plus(std_gt_vcf)[0],
config=data["config"])
gt_vcfs[dd.get_sample_name(data)] = _filter_by_support(gt_vcf, data)
if paired and paired.normal_name:
gt_vcfs = _filter_by_background([paired.tumor_name], [paired.normal_name], gt_vcfs, paired.tumor_data)
out = []
for data in items:
if "sv" not in data:
data["sv"] = []
vcf_file = gt_vcfs[dd.get_sample_name(data)]
if dd.get_svprioritize(data):
effects_vcf, _ = effects.add_to_vcf(vcf_file, data, "snpeff")
else:
effects_vcf = None
data["sv"].append({"variantcaller": "lumpy",
"vrn_file": effects_vcf or vcf_file,
"exclude_file": exclude_file})
out.append(data)
return out
def _run_cnvkit_cancer(items, background, access_file):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(items[0])
ckout = _run_cnvkit_shared(items[0], [paired.tumor_bam], [paired.normal_bam],
access_file, work_dir, background_name=paired.normal_name)
ckout = theta.run(ckout, paired)
return _associate_cnvkit_out(ckout, items)
def _run_cnvkit_cancer(items, background, access_file):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(items[0])
ckout = _run_cnvkit_shared(items[0], [paired.tumor_bam], [paired.normal_bam],
access_file, work_dir, background_name=paired.normal_name)
ckout = theta.run(ckout, paired)
return _associate_cnvkit_out(ckout, items)
def run(items):
"""Perform detection of structural variations with lumpy, using bwa-mem alignment.
"""
if not all(utils.get_in(data, ("config", "algorithm", "aligner"))
in ["bwa", "sentieon-bwa", False, None] for data in items):
raise ValueError("Require bwa-mem alignment input for lumpy structural variation detection")
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data if paired and paired.tumor_data else items[0])
previous_evidence = {}
full_bams, sr_bams, disc_bams = [], [], []
for data in items:
sr_bam, disc_bam = sshared.get_split_discordants(data, work_dir)
full_bams.append(dd.get_align_bam(data))
sr_bams.append(sr_bam)
disc_bams.append(disc_bam)
cur_dels, cur_dups = _bedpes_from_cnv_caller(data, work_dir)
previous_evidence[dd.get_sample_name(data)] = {}
if cur_dels and utils.file_exists(cur_dels):
previous_evidence[dd.get_sample_name(data)]["dels"] = cur_dels
if cur_dups and utils.file_exists(cur_dups):
previous_evidence[dd.get_sample_name(data)]["dups"] = cur_dups
lumpy_vcf, exclude_file = _run_lumpy(full_bams, sr_bams, disc_bams, previous_evidence,
work_dir, items)
gt_vcfs = {}
for data in items:
sample = dd.get_sample_name(data)
sample_vcf = vcfutils.select_sample(lumpy_vcf, sample,
utils.append_stem(lumpy_vcf, "-%s" % sample),
data["config"])
if "bnd-genotype" in dd.get_tools_on(data):
gt_vcf = _run_svtyper(sample_vcf, dd.get_align_bam(data), exclude_file, data)
else:
std_vcf, bnd_vcf = _split_breakends(sample_vcf, data)
std_gt_vcf = _run_svtyper(std_vcf, dd.get_align_bam(data), exclude_file, data)
gt_vcf = vcfutils.concat_variant_files_bcftools(
orig_files=[std_gt_vcf, bnd_vcf],
out_file="%s-combined.vcf.gz" % utils.splitext_plus(std_gt_vcf)[0],
config=data["config"])
gt_vcfs[dd.get_sample_name(data)] = _filter_by_support(gt_vcf, data)
if paired and paired.normal_name:
gt_vcfs = _filter_by_background([paired.tumor_name], [paired.normal_name], gt_vcfs, paired.tumor_data)
out = []
for data in items:
if "sv" not in data:
data["sv"] = []
vcf_file = gt_vcfs[dd.get_sample_name(data)]
if dd.get_svprioritize(data):
effects_vcf, _ = effects.add_to_vcf(vcf_file, data, "snpeff")
else:
effects_vcf = None
data["sv"].append({"variantcaller": "lumpy",
"vrn_file": effects_vcf or vcf_file,
"exclude_file": exclude_file})
out.append(data)
return out
def run(items):
"""Perform detection of structural variations with lumpy, using bwa-mem alignment.
"""
if not all(
utils.get_in(data, ("config", "algorithm",
"aligner")) in ["bwa", False, None]
for data in items):
raise ValueError(
"Require bwa-mem alignment input for lumpy structural variation detection"
)
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(
paired.tumor_data if paired and paired.tumor_data else items[0])
full_bams, sr_bams, disc_bams = [], [], []
for data in items:
dedup_bam, sr_bam, disc_bam = sshared.get_split_discordants(
data, work_dir)
full_bams.append(dedup_bam)
sr_bams.append(sr_bam)
disc_bams.append(disc_bam)
lumpy_vcf, exclude_file = _run_lumpy(full_bams, sr_bams, disc_bams,
work_dir, items)
gt_vcfs = {}
for data in items:
sample = dd.get_sample_name(data)
dedup_bam, sr_bam, _ = sshared.get_split_discordants(data, work_dir)
sample_vcf = vcfutils.select_sample(
lumpy_vcf, sample, utils.append_stem(lumpy_vcf, "-%s" % sample),
data["config"])
std_vcf, bnd_vcf = _split_breakends(sample_vcf, data)
std_gt_vcf = _run_svtyper(std_vcf, dedup_bam, sr_bam, exclude_file,
data)
gt_vcf = vcfutils.combine_variant_files(
orig_files=[std_gt_vcf, bnd_vcf],
out_file="%s-combined.vcf.gz" % utils.splitext_plus(std_gt_vcf)[0],
ref_file=dd.get_ref_file(data),
config=data["config"])
gt_vcfs[dd.get_sample_name(data)] = _filter_by_support(gt_vcf, data)
if paired and paired.normal_name:
gt_vcfs = _filter_by_background([paired.tumor_name],
[paired.normal_name], gt_vcfs,
paired.tumor_data)
out = []
for data in items:
if "sv" not in data:
data["sv"] = []
vcf_file = gt_vcfs[dd.get_sample_name(data)]
effects_vcf, _ = effects.add_to_vcf(vcf_file, data, "snpeff")
data["sv"].append({
"variantcaller": "lumpy",
"vrn_file": effects_vcf or vcf_file,
"exclude_file": exclude_file
})
out.append(data)
return out
def _run_vardict_paired(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Detect variants with Vardict.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(
align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(items[0], out_file) as tx_out_file:
paired = vcfutils.get_paired_bams(align_bams, items)
if not paired.normal_bam:
ann_file = _run_vardict_caller(align_bams, items, ref_file,
assoc_files, region, out_file)
return ann_file
vcffilter = config_utils.get_program("vcffilter", config)
vardict = config_utils.get_program("vardict", config)
vcfstreamsort = config_utils.get_program("vcfstreamsort", config)
vcfallelicprimitives = config_utils.get_program(
"vcfallelicprimitives", config)
strandbias = "testsomatic.R"
var2vcf = "var2vcf_somatic.pl"
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
freq = float(
utils.get_in(config,
("algorithm", "min_allele_fraction"), 10)) / 100.0
opts = " ".join(
_vardict_options_from_config(items, config, out_file, region))
coverage_interval = utils.get_in(
config, ("algorithm", "coverage_interval"), "exome")
# for deep targeted panels, require 50 worth of coverage
var2vcf_opts = " -v 50 " if coverage_interval == "regional" else ""
fix_ambig = vcfutils.fix_ambiguous_cl()
cmd = (
"{vardict} -G {ref_file} -f {freq} "
"-N {paired.tumor_name} -b \"{paired.tumor_bam}|{paired.normal_bam}\" {opts} "
"| {strandbias} "
"| {var2vcf} -N \"{paired.tumor_name}|{paired.normal_name}\" -f {freq} {var2vcf_opts} "
"| {fix_ambig} | {vcfstreamsort} {compress_cmd} > {tx_out_file}"
)
bam.index(paired.tumor_bam, config)
bam.index(paired.normal_bam, config)
do.run(cmd.format(**locals()),
"Genotyping with VarDict: Inference", {})
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"),
ref_file, config)
return ann_file
def _ready_for_het_analysis(items):
"""Check if a sample has input information for heterogeneity analysis.
We currently require a tumor/normal sample containing both CNV and variant calls.
"""
paired = vcfutils.get_paired_bams([dd.get_align_bam(d) for d in items],
items)
has_het = any(dd.get_hetcaller(d) for d in items)
if has_het and paired:
return get_variants(paired.tumor_data) and _get_calls(
paired.tumor_data, cnv_only=True)
def _run_cnvkit_cancer(items, background):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
normal_data = [x for x in items if dd.get_sample_name(x) != paired.tumor_name]
ckouts = _run_cnvkit_shared([paired.tumor_data], normal_data)
if not ckouts:
return items
assert len(ckouts) == 1
tumor_data = _associate_cnvkit_out(ckouts, [paired.tumor_data], is_somatic=True)
return tumor_data + normal_data
def _run_freebayes_paired(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Detect SNPs and indels with FreeBayes.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(items[0], out_file) as tx_out_file:
paired = get_paired_bams(align_bams, items)
if not paired.normal_bam:
return _run_freebayes_caller(align_bams, items, ref_file,
assoc_files, region, out_file)
#raise ValueError("Require both tumor and normal BAM files for FreeBayes cancer calling")
vcfsamplediff = config_utils.get_program("vcfsamplediff", config)
vcffilter = config_utils.get_program("vcffilter", config)
vcfallelicprimitives = config_utils.get_program("vcfallelicprimitives", config)
vcfstreamsort = config_utils.get_program("vcfstreamsort", config)
freebayes = config_utils.get_program("freebayes", config)
opts = " ".join(_freebayes_options_from_config(items, config, out_file, region))
if "--min-alternate-fraction" not in opts and "-F" not in opts:
# add minimum reportable allele frequency
# FreeBayes defaults to 20%, but use 10% by default for the
# tumor case
min_af = float(utils.get_in(paired.tumor_config, ("algorithm",
"min_allele_fraction"), 10)) / 100.0
opts += " --min-alternate-fraction %s" % min_af
opts += " --min-repeat-entropy 1 --experimental-gls"
# Recommended settings for cancer calling
# https://groups.google.com/d/msg/freebayes/dTWBtLyM4Vs/HAK_ZhJHguMJ
opts += " --pooled-discrete --genotype-qualities --report-genotype-likelihood-max"
# NOTE: The first sample name in the vcfsamplediff call is
# the one supposed to be the *germline* one
# NOTE: -s in vcfsamplediff (strict checking: i.e., require no
# reads in the germline to call somatic) is not used as it is
# too stringent
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
cl = ("{freebayes} -f {ref_file} {opts} "
"{paired.tumor_bam} {paired.normal_bam} "
"| {vcffilter} -f 'QUAL > 5' -s "
"| {vcfallelicprimitives} | {vcfstreamsort} "
"| {vcfsamplediff} VT {paired.normal_name} {paired.tumor_name} - "
"{compress_cmd} > {tx_out_file}")
bam.index(paired.tumor_bam, config)
bam.index(paired.normal_bam, config)
do.run(cl.format(**locals()), "Genotyping paired variants with FreeBayes", {})
fix_somatic_calls(out_file, config)
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"), ref_file,
config)
return ann_file
def _run_cnvkit_cancer(items, background):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
normal_data = [x for x in items if dd.get_sample_name(x) != paired.tumor_name]
ckouts = _run_cnvkit_shared([paired.tumor_data], normal_data)
if not ckouts:
return items
assert len(ckouts) == 1
tumor_data = _associate_cnvkit_out(ckouts, [paired.tumor_data], is_somatic=True)
return tumor_data + normal_data
def _paired_load_script(work_bams, names, chrom, pairmode, items):
"""Prepare BAMs for assessing CNVs in a paired tumor/normal setup.
"""
paired = vcfutils.get_paired_bams(work_bams, items)
return _paired_prep.format(case_file=paired.tumor_bam,
case_name=paired.tumor_name,
ctrl_file=paired.normal_bam,
ctrl_name=paired.normal_name,
num_cores=0,
chrom=chrom,
pairmode=pairmode)
def run(align_bams, items, ref_file, assoc_files, region, out_file):
"""Run octopus variant calling, handling both somatic and germline calling.
"""
if not utils.file_exists(out_file):
paired = vcfutils.get_paired_bams(align_bams, items)
regions = _get_regions(region, out_file, items)
if paired:
return _run_somatic(paired, ref_file, regions, out_file)
else:
return _run_germline(align_bams, items, ref_file, regions,
out_file)
return out_file
def run(align_bams, items, ref_file, assoc_files, region=None, out_file=None):
"""Run strelka2 variant calling, either paired tumor/normal or germline calling.
"""
if vcfutils.is_paired_analysis(align_bams, items):
paired = vcfutils.get_paired_bams(align_bams, items)
assert paired.normal_bam, "Strelka2 requires a normal sample"
call_file = _run_somatic(paired, ref_file, assoc_files, region,
out_file)
else:
call_file = _run_germline(align_bams, items, ref_file, assoc_files,
region, out_file)
return call_file
def _run_freebayes_paired(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Detect SNPs and indels with FreeBayes.
This is used for paired tumor / normal samples. Sources of options for FreeBayes:
mailing list: https://groups.google.com/d/msg/freebayes/dTWBtLyM4Vs/HAK_ZhJHguMJ
mailing list: https://groups.google.com/forum/#!msg/freebayes/LLH7ZfZlVNs/63FdD31rrfEJ
speedseq: https://github.com/cc2qe/speedseq/blob/e6729aa2589eca4e3a946f398c1a2bdc15a7300d/bin/speedseq#L916
sga/freebayes: https://github.com/jts/sga-extra/blob/7e28caf71e8107b697f9be7162050e4fa259694b/
sga_generate_varcall_makefile.pl#L299
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(items[0], out_file) as tx_out_file:
paired = get_paired_bams(align_bams, items)
if not paired.normal_bam:
return _run_freebayes_caller(align_bams, items, ref_file,
assoc_files, region, out_file)
#raise ValueError("Require both tumor and normal BAM files for FreeBayes cancer calling")
freebayes = config_utils.get_program("freebayes", config)
opts = " ".join(_freebayes_options_from_config(items, config, out_file, region))
if "--min-alternate-fraction" not in opts and "-F" not in opts:
# add minimum reportable allele frequency
# FreeBayes defaults to 20%, but use 10% by default for the
# tumor case
min_af = float(utils.get_in(paired.tumor_config, ("algorithm",
"min_allele_fraction"), 10)) / 100.0
opts += " --min-alternate-fraction %s" % min_af
opts += " --min-repeat-entropy 1 --experimental-gls"
# Recommended settings for cancer calling
opts += (" --pooled-discrete --pooled-continuous --genotype-qualities "
"--report-genotype-likelihood-max --allele-balance-priors-off")
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
fix_ambig = vcfutils.fix_ambiguous_cl()
py_cl = os.path.join(os.path.dirname(sys.executable), "py")
cl = ("{freebayes} -f {ref_file} {opts} "
"{paired.tumor_bam} {paired.normal_bam} "
"| vcffilter -f 'QUAL > 5' -s "
"| {py_cl} -x 'bcbio.variation.freebayes.call_somatic(x)' "
"| {fix_ambig} | vcfallelicprimitives --keep-info --keep-geno "
"| vt normalize -q -r {ref_file} - "
"{compress_cmd} > {tx_out_file}")
bam.index(paired.tumor_bam, config)
bam.index(paired.normal_bam, config)
do.run(cl.format(**locals()), "Genotyping paired variants with FreeBayes", {})
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"), ref_file,
config)
return ann_file
def _paired_load_script(work_bams, names, chrom, pairmode, items):
"""Prepare BAMs for assessing CNVs in a paired tumor/normal setup.
"""
paired = vcfutils.get_paired_bams(work_bams, items)
bed_file = _get_regional_bed_file(items[0])
if bed_file:
return _paired_prep_targeted.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode, bed_file=bed_file)
else:
return _paired_prep.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode)
def run(align_bams, items, ref_file, assoc_files, region, out_file):
"""Run strelka2 variant calling, either paired tumor/normal or germline calling.
region can be a single region or list of multiple regions for multicore calling.
"""
if vcfutils.is_paired_analysis(align_bams, items):
paired = vcfutils.get_paired_bams(align_bams, items)
assert paired.normal_bam, "Strelka2 requires a normal sample"
call_file = _run_somatic(paired, ref_file, assoc_files, region, out_file)
else:
call_file = _run_germline(align_bams, items, ref_file,
assoc_files, region, out_file)
return call_file
def run(align_bams, items, ref_file, assoc_files, region, out_file):
"""Run octopus variant calling, handling both somatic and germline calling.
"""
if not utils.file_exists(out_file):
paired = vcfutils.get_paired_bams(align_bams, items)
vrs = bedutils.population_variant_regions(items)
target = shared.subset_variant_regions(vrs, region,
out_file, items=items, do_merge=True)
if paired:
return _run_somatic(paired, ref_file, target, out_file)
else:
return _run_germline(align_bams, items, ref_file, target, out_file)
return out_file
def _run_cnvkit_cancer(items, background):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data)
ckouts = _run_cnvkit_shared([paired.tumor_data], [paired.tumor_bam], [paired.normal_bam],
work_dir, background_name=paired.normal_name)
if not ckouts:
return items
assert len(ckouts) == 1
tumor_data = _associate_cnvkit_out(ckouts, [paired.tumor_data])
normal_data = [x for x in items if dd.get_sample_name(x) != paired.tumor_name]
return tumor_data + normal_data
def _paired_load_script(work_bams, names, chrom, pairmode, items):
"""Prepare BAMs for assessing CNVs in a paired tumor/normal setup.
"""
paired = vcfutils.get_paired_bams(work_bams, items)
bed_file = _get_regional_bed_file(items[0])
if bed_file:
return _paired_prep_targeted.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode, bed_file=bed_file)
else:
return _paired_prep.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode)
def _run_cnvkit_cancer(items, background):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data)
access_file = _create_access_file(dd.get_ref_file(paired.tumor_data), work_dir, paired.tumor_data)
ckout = _run_cnvkit_shared(paired.tumor_data, [paired.tumor_bam], [paired.normal_bam],
access_file, work_dir, background_name=paired.normal_name)
# Skip THetA runs until we can speed up data preparation steps
# ckout = theta.run(ckout, paired)
tumor_data = _associate_cnvkit_out(ckout, [paired.tumor_data])
normal_data = [x for x in items if dd.get_sample_name(x) != paired.tumor_name]
return tumor_data + normal_data
def _run_cnvkit_cancer(items, background):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data)
access_file = _create_access_file(dd.get_ref_file(paired.tumor_data), work_dir, paired.tumor_data)
ckout = _run_cnvkit_shared(paired.tumor_data, [paired.tumor_bam], [paired.normal_bam],
access_file, work_dir, background_name=paired.normal_name)
if not ckout:
return items
tumor_data = _associate_cnvkit_out(ckout, [paired.tumor_data])
normal_data = [x for x in items if dd.get_sample_name(x) != paired.tumor_name]
return tumor_data + normal_data
def _paired_load_script(work_bams, names, chrom, pairmode, items):
"""Prepare BAMs for assessing CNVs in a paired tumor/normal setup.
"""
paired = vcfutils.get_paired_bams(work_bams, items)
bed_file = items[0]["config"]["algorithm"].get("variant_regions", None)
is_genome = items[0]["config"]["algorithm"].get("coverage_interval", "exome").lower() in ["genome"]
if utils.file_exists(bed_file) and not is_genome:
return _paired_prep_targeted.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode, bed_file=bed_file)
else:
return _paired_prep.format(case_file=paired.tumor_bam, case_name=paired.tumor_name,
ctrl_file=paired.normal_bam, ctrl_name=paired.normal_name,
num_cores=0, chrom=chrom, pairmode=pairmode)
def run_varscan(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
paired = get_paired_bams(align_bams, items)
if paired and paired.normal_bam and paired.tumor_bam:
call_file = samtools.shared_variantcall(_varscan_paired, "varscan",
align_bams, ref_file, items,
assoc_files, region, out_file)
else:
vcfutils.check_paired_problems(items)
call_file = samtools.shared_variantcall(_varscan_work, "varscan",
align_bams, ref_file,
items, assoc_files,
region, out_file)
return call_file
def mutect2_caller(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Call variation with GATK's MuTect2.
This requires the full non open-source version of GATK 3.5+.
"""
if out_file is None:
out_file = "%s-variants.vcf.gz" % utils.splitext_plus(align_bams[0])[0]
if not utils.file_exists(out_file):
paired = vcfutils.get_paired_bams(align_bams, items)
broad_runner = broad.runner_from_config(items[0]["config"])
gatk_type = broad_runner.gatk_type()
_prep_inputs(align_bams, ref_file, items)
with file_transaction(items[0], out_file) as tx_out_file:
params = ["-T", "Mutect2" if gatk_type == "gatk4" else "MuTect2",
"--annotation", "ClippingRankSumTest",
"--annotation", "DepthPerSampleHC"]
if gatk_type == "gatk4":
params += ["--reference", ref_file]
else:
params += ["-R", ref_file]
for a in annotation.get_gatk_annotations(items[0]["config"], include_baseqranksum=False):
params += ["--annotation", a]
# Avoid issues with BAM CIGAR reads that GATK doesn't like
if gatk_type == "gatk4":
params += ["--read-validation-stringency", "LENIENT"]
params += _add_tumor_params(paired, items, gatk_type)
params += _add_region_params(region, out_file, items, gatk_type)
# Avoid adding dbSNP/Cosmic so they do not get fed to variant filtering algorithm
# Not yet clear how this helps or hurts in a general case.
#params += _add_assoc_params(assoc_files)
resources = config_utils.get_resources("mutect2", items[0]["config"])
if "options" in resources:
params += [str(x) for x in resources.get("options", [])]
assert LooseVersion(broad_runner.gatk_major_version()) >= LooseVersion("3.5"), \
"Require full version of GATK 3.5+ for mutect2 calling"
broad_runner.new_resources("mutect2")
gatk_cmd = broad_runner.cl_gatk(params, os.path.dirname(tx_out_file))
if gatk_type == "gatk4":
tx_raw_prefilt_file = "%s-raw%s" % utils.splitext_plus(tx_out_file)
tx_raw_file = "%s-raw-filt%s" % utils.splitext_plus(tx_out_file)
filter_cmd = _mutect2_filter(broad_runner, tx_raw_prefilt_file, tx_raw_file, ref_file)
cmd = "{gatk_cmd} -O {tx_raw_prefilt_file} && {filter_cmd}"
else:
tx_raw_file = "%s-raw%s" % utils.splitext_plus(tx_out_file)
cmd = "{gatk_cmd} > {tx_raw_file}"
do.run(cmd.format(**locals()), "MuTect2")
out_file = _af_filter(paired.tumor_data, tx_raw_file, out_file)
return vcfutils.bgzip_and_index(out_file, items[0]["config"])
def estimate(items, batch, config):
"""Estimate heterogeneity for a pair of tumor/normal samples. Run in parallel.
XXX In progress, currently uses THetA but not yet turned on
"""
paired = vcfutils.get_paired_bams([dd.get_align_bam(d) for d in items], items)
cnvs = _get_cnvs(paired.tumor_data)
new_cnvs = theta.run(cnvs[0], paired)
print(new_cnvs)
out = []
for data in items:
if batch == _get_batches(data)[0]:
out.append([data])
return out
def run(align_bams, items, ref_file, assoc_files, region, out_file):
"""Run strelka2 variant calling, either paired tumor/normal or germline calling.
region can be a single region or list of multiple regions for multicore calling.
"""
call_file = "%s-raw.vcf.gz" % utils.splitext_plus(out_file)[0]
strelka_work_dir = "%s-work" % utils.splitext_plus(out_file)[0]
paired = vcfutils.get_paired_bams(align_bams, items)
if paired:
assert paired.normal_bam, "Strelka2 requires a normal sample"
call_file = _run_somatic(paired, ref_file, assoc_files, region, call_file, strelka_work_dir)
else:
call_file = _run_germline(align_bams, items, ref_file,
assoc_files, region, call_file, strelka_work_dir)
return _af_annotate_and_filter(paired, items, call_file, out_file)
def run(align_bams, items, ref_file, assoc_files, region, out_file):
"""Run strelka2 variant calling, either paired tumor/normal or germline calling.
region can be a single region or list of multiple regions for multicore calling.
"""
call_file = "%s-raw.vcf.gz" % utils.splitext_plus(out_file)[0]
strelka_work_dir = "%s-work" % utils.splitext_plus(out_file)[0]
paired = vcfutils.get_paired_bams(align_bams, items)
if paired:
assert paired.normal_bam, "Strelka2 requires a normal sample"
call_file = _run_somatic(paired, ref_file, assoc_files, region, call_file, strelka_work_dir)
else:
call_file = _run_germline(align_bams, items, ref_file,
assoc_files, region, call_file, strelka_work_dir)
return _af_annotate_and_filter(paired, items, call_file, out_file)
def _run_tumor_pindel_caller(align_bams,
items,
ref_file,
assoc_files,
region=None,
out_file=None):
"""Detect indels with pindel in tumor/[normal] analysis.
Only attempts to detect small insertion/deletions and not larger structural events.
:param align_bam: (list) bam files
:param items: (dict) information from yaml
:param ref_file: (str) genome in fasta format
:param assoc_file: (dict) files for annotation
:param region: (str or tupple) region to analyze
:param out_file: (str) final vcf file
:returns: (str) final vcf file
"""
config = items[0]["config"]
paired = get_paired_bams(align_bams, items)
if out_file is None:
out_file = "%s-indels.vcf" % os.path.splitext(align_bams[0])[0]
paired_bam = [paired.tumor_bam]
paired_name = [paired.tumor_name]
if paired.normal_bam:
paired_bam.append(paired.normal_bam)
paired_name.append(paired.normal_name)
if not utils.file_exists(out_file):
with tx_tmpdir(config) as tmp_path:
for align_bam in align_bams:
bam.index(align_bam, config)
root_pindel = os.path.join(tmp_path, "pindelroot")
pindel = config_utils.get_program("pindel", config)
opts = _pindel_options(items, config, out_file, region, tmp_path)
tmp_input = _create_tmp_input(paired_bam, paired_name, tmp_path,
config)
cmd = (
"{pindel} -f {ref_file} -i {tmp_input} -o {root_pindel} " +
"{opts} --report_inversions false --report_duplications false "
"--report_long_insertions false --report_breakpoints false "
"--report_interchromosomal_events false "
"--max_range_index 2")
do.run(cmd.format(**locals()), "Genotyping with pindel", {})
out_file = _create_vcf(root_pindel, out_file, ref_file, items,
paired)
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"),
ref_file, config)
return ann_file
def _run_vardict_paired(align_bams, items, ref_file, assoc_files,
region=None, out_file=None):
"""Detect variants with Vardict.
This is used for paired tumor / normal samples.
"""
config = items[0]["config"]
if out_file is None:
out_file = "%s-paired-variants.vcf.gz" % os.path.splitext(align_bams[0])[0]
if not utils.file_exists(out_file):
with file_transaction(items[0], out_file) as tx_out_file:
paired = vcfutils.get_paired_bams(align_bams, items)
if not paired.normal_bam:
ann_file = _run_vardict_caller(align_bams, items, ref_file,
assoc_files, region, out_file)
return ann_file
vcffilter = config_utils.get_program("vcffilter", config)
vardict = config_utils.get_program("vardict", config)
vcfstreamsort = config_utils.get_program("vcfstreamsort", config)
strandbias = "testsomatic.R"
var2vcf = "var2vcf_paired.pl"
compress_cmd = "| bgzip -c" if out_file.endswith("gz") else ""
freq = float(utils.get_in(config, ("algorithm", "min_allele_fraction"), 10)) / 100.0
# merge bed file regions as amplicon VarDict is only supported in single sample mode
opts = " ".join(_vardict_options_from_config(items, config, out_file, region, do_merge=True))
coverage_interval = utils.get_in(config, ("algorithm", "coverage_interval"), "exome")
# for deep targeted panels, require 50 worth of coverage
var2vcf_opts = " -v 50 " if coverage_interval == "regional" else ""
fix_ambig = vcfutils.fix_ambiguous_cl()
if any("vardict_somatic_filter" in tz.get_in(("config", "algorithm", "tools_off"), data, [])
for data in items):
somatic_filter = ""
else:
somatic_filter = ("| %s -x 'bcbio.variation.freebayes.call_somatic(x)'" %
os.path.join(os.path.dirname(sys.executable), "py"))
cmd = ("{vardict} -G {ref_file} -f {freq} "
"-N {paired.tumor_name} -b \"{paired.tumor_bam}|{paired.normal_bam}\" {opts} "
"| {strandbias} "
"| {var2vcf} -M -N \"{paired.tumor_name}|{paired.normal_name}\" -f {freq} {var2vcf_opts} "
"{somatic_filter} | {fix_ambig} | {vcfstreamsort} {compress_cmd} > {tx_out_file}")
bam.index(paired.tumor_bam, config)
bam.index(paired.normal_bam, config)
do.run(cmd.format(**locals()), "Genotyping with VarDict: Inference", {})
ann_file = annotation.annotate_nongatk_vcf(out_file, align_bams,
assoc_files.get("dbsnp"), ref_file,
config)
return ann_file
def _run_cnvkit_cancer(items, background):
"""Run CNVkit on a tumor/normal pair.
"""
paired = vcfutils.get_paired_bams([x["align_bam"] for x in items], items)
work_dir = _sv_workdir(paired.tumor_data)
ckouts = _run_cnvkit_shared([paired.tumor_data], [paired.tumor_bam],
[paired.normal_bam],
work_dir,
background_name=paired.normal_name)
if not ckouts:
return items
assert len(ckouts) == 1
tumor_data = _associate_cnvkit_out(ckouts, [paired.tumor_data])
normal_data = [
x for x in items if dd.get_sample_name(x) != paired.tumor_name
]
return tumor_data + normal_data
def run(align_bams, items, ref_file, assoc_files, region=None, out_file=None):
"""Run tumor only smCounter2 calling.
"""
paired = vcfutils.get_paired_bams(align_bams, items)
assert paired and not paired.normal_bam, (
"smCounter2 supports tumor-only variant calling: %s" %
(",".join([dd.get_sample_name(d) for d in items])))
vrs = bedutils.population_variant_regions(items)
target = shared.subset_variant_regions(vrs,
region,
out_file,
items=items,
do_merge=True)
out_file = out_file.replace(".vcf.gz", ".vcf")
out_prefix = utils.splitext_plus(os.path.basename(out_file))[0]
if not utils.file_exists(out_file) and not utils.file_exists(out_file +
".gz"):
with file_transaction(paired.tumor_data, out_file) as tx_out_file:
cmd = [
"smCounter2", "--runPath",
os.path.dirname(tx_out_file), "--outPrefix", out_prefix,
"--bedTarget", target, "--refGenome", ref_file, "--bamFile",
paired.tumor_bam, "--bamType", "consensus", "--nCPU",
dd.get_num_cores(paired.tumor_data)
]
do.run(cmd, "smcounter2 variant calling")
for fname in glob.glob(
os.path.join(os.path.dirname(tx_out_file),
"*.smCounter*")):
shutil.move(
fname,
os.path.join(os.path.dirname(out_file),
os.path.basename(fname)))
utils.symlink_plus(
os.path.join(os.path.dirname(out_file),
"%s.smCounter.cut.vcf" % out_prefix), out_file)
return vcfutils.bgzip_and_index(
out_file,
paired.tumor_data["config"],
remove_orig=False,
prep_cmd="sed 's#FORMAT\t%s#FORMAT\t%s#' | %s" %
(out_prefix, dd.get_sample_name(paired.tumor_data),
vcfutils.add_contig_to_header_cl(dd.get_ref_file(paired.tumor_data),
out_file)))