def variants(data, out_dir):
"""Variants QC metrics"""
if not "variants" in data:
return None
work_dir = safe_makedir(out_dir)
sample = dd.get_sample_name(data)
bcfstats = _run_bcftools(data, work_dir)
bed_file = dd.get_coverage(data)
bcf_out = os.path.join(sample + "_bcbio_variants_stats.txt")
cg_file = os.path.join(sample + "_with-gc.vcf.gz")
parse_file = os.path.join(sample + "_gc-depth-parse.tsv")
qc_file = os.path.join(sample + "_bcbio_variants.txt")
with chdir(work_dir):
if not file_exists(bcf_out):
with open(bcf_out, "w") as out_handle:
yaml.safe_dump(bcfstats,
out_handle,
default_flow_style=False,
allow_unicode=False)
if "vrn_file" not in data or not bed_file:
return None
in_vcf = data['vrn_file']
cleaned_bed = clean_file(bed_file, data)
if file_exists(qc_file):
return qc_file
in_bam = dd.get_align_bam(data) or dd.get_work_bam(data)
ref_file = dd.get_ref_file(data)
assert ref_file, "Need the reference genome fasta file."
bed_file = dd.get_variant_regions(data)
in_bam = dd.get_align_bam(data) or dd.get_work_bam(data)
num_cores = dd.get_num_cores(data)
broad_runner = broad.runner_from_config_safe(data["config"])
if in_bam and broad_runner and broad_runner.has_gatk():
if not file_exists(parse_file):
with file_transaction(cg_file) as tx_out:
params = [
"-T", "VariantAnnotator", "-R", ref_file, "-L",
cleaned_bed, "-I", in_bam, "-A", "GCContent", "-A",
"Coverage", "--variant", in_vcf, "--out", tx_out
]
broad_runner.run_gatk(params)
cg_file = vcfutils.bgzip_and_index(cg_file, data["config"])
if not file_exists(parse_file):
with file_transaction(parse_file) as out_tx:
with open(out_tx, 'w') as out_handle:
print >> out_handle, "CG\tdepth\tsample"
cmd = (
"bcftools query -s {sample} -f '[%GC][\\t%DP][\\t%SAMPLE]\\n' -R "
"{bed_file} {cg_file} >> {out_tx}")
do.run(cmd.format(**locals()),
"Calculating GC content and depth for %s" % in_vcf)
logger.debug('parsing coverage: %s' % sample)
if not file_exists(qc_file):
# This files will be copied to final
_summary_variants(parse_file, qc_file)
if file_exists(qc_file) and file_exists(parse_file):
remove_plus(cg_file)
def variants(data, out_dir):
"""Variants QC metrics"""
if not "variants" in data:
return None
work_dir = safe_makedir(out_dir)
sample = dd.get_sample_name(data)
bcfstats = _run_bcftools(data, work_dir)
bed_file = dd.get_coverage(data)
bcf_out = os.path.join(sample + "_bcbio_variants_stats.txt")
cg_file = os.path.join(sample + "_with-gc.vcf.gz")
parse_file = os.path.join(sample + "_gc-depth-parse.tsv")
qc_file = os.path.join(sample + "_bcbio_variants.txt")
with chdir(work_dir):
if not file_exists(bcf_out):
with open(bcf_out, "w") as out_handle:
yaml.safe_dump(bcfstats, out_handle, default_flow_style=False, allow_unicode=False)
if "vrn_file" not in data or not bed_file:
return None
in_vcf = data['vrn_file']
cleaned_bed = clean_file(bed_file, data)
if file_exists(qc_file):
return qc_file
in_bam = dd.get_align_bam(data) or dd.get_work_bam(data)
ref_file = dd.get_ref_file(data)
assert ref_file, "Need the reference genome fasta file."
bed_file = dd.get_variant_regions(data)
in_bam = dd.get_align_bam(data) or dd.get_work_bam(data)
num_cores = dd.get_num_cores(data)
broad_runner = broad.runner_from_config_safe(data["config"])
if in_bam and broad_runner and broad_runner.has_gatk():
if not file_exists(parse_file):
with file_transaction(cg_file) as tx_out:
params = ["-T", "VariantAnnotator",
"-R", ref_file,
"-L", cleaned_bed,
"-I", in_bam,
"-A", "GCContent",
"-A", "Coverage",
"--variant", in_vcf,
"--out", tx_out]
broad_runner.run_gatk(params)
cg_file = vcfutils.bgzip_and_index(cg_file, data["config"])
if not file_exists(parse_file):
with file_transaction(parse_file) as out_tx:
with open(out_tx, 'w') as out_handle:
print >>out_handle, "CG\tdepth\tsample"
cmd = ("bcftools query -s {sample} -f '[%GC][\\t%DP][\\t%SAMPLE]\\n' -R "
"{bed_file} {cg_file} >> {out_tx}")
do.run(cmd.format(**locals()),
"Calculating GC content and depth for %s" % in_vcf)
logger.debug('parsing coverage: %s' % sample)
if not file_exists(qc_file):
# This files will be copied to final
_summary_variants(parse_file, qc_file)
if file_exists(qc_file) and file_exists(parse_file):
remove_plus(cg_file)
def _run_germline(align_bams, items, ref_file, assoc_files, region, out_file, work_dir):
if not utils.file_exists(out_file):
with file_transaction(items[0], work_dir) as tx_work_dir:
workflow_file = _configure_germline(align_bams, items, ref_file, region, out_file, tx_work_dir)
_run_workflow(items[0], workflow_file, tx_work_dir)
raw_file = os.path.join(work_dir, "results", "variants",
"genome.vcf.gz" if joint.want_gvcf(items) else "variants.vcf.gz")
utils.copy_plus(raw_file, out_file)
# Remove files with relative symlinks
utils.remove_plus(os.path.join(work_dir, "results", "variants", "genome.vcf.gz"))
return vcfutils.bgzip_and_index(out_file, items[0]["config"])
def _run_germline(align_bams, items, ref_file, assoc_files, region, out_file, work_dir):
if not utils.file_exists(out_file):
with file_transaction(items[0], work_dir) as tx_work_dir:
workflow_file = _configure_germline(align_bams, items, ref_file, region, out_file, tx_work_dir)
_run_workflow(items[0], workflow_file, tx_work_dir)
raw_file = os.path.join(work_dir, "results", "variants",
"genome.vcf.gz" if joint.want_gvcf(items) else "variants.vcf.gz")
utils.copy_plus(raw_file, out_file)
# Remove files with relative symlinks
utils.remove_plus(os.path.join(work_dir, "results", "variants", "genome.vcf.gz"))
return vcfutils.bgzip_and_index(out_file, items[0]["config"])
def _run_germline(align_bams, items, ref_file, assoc_files, region, out_file,
work_dir):
if not utils.file_exists(out_file):
with file_transaction(items[0], work_dir) as tx_work_dir:
workflow_file = _configure_germline(align_bams, items, ref_file,
region, out_file, tx_work_dir)
if workflow_file:
has_variants = True
_run_workflow(items[0], workflow_file, tx_work_dir)
else:
has_variants = False
vcfutils.write_empty_vcf(
out_file, items[0]["config"],
[dd.get_sample_name(d) for d in items])
if has_variants:
raw_file = os.path.join(
work_dir, "results", "variants", "genome.vcf.gz"
if joint.want_gvcf(items) else "variants.vcf.gz")
utils.copy_plus(raw_file, out_file)
# Remove files with relative symlinks
utils.remove_plus(
os.path.join(work_dir, "results", "variants", "genome.vcf.gz"))
return vcfutils.bgzip_and_index(out_file, items[0]["config"])