def _add_variantcalls_to_output(out, data):
"""Call ploidy and convert into VCF and BED representations.
"""
call_file = "%s-call%s" % os.path.splitext(out["cns"])
gender = dd.get_gender(data)
if not utils.file_exists(call_file):
with file_transaction(data, call_file) as tx_call_file:
cmd = [os.path.join(os.path.dirname(sys.executable), "cnvkit.py"), "call",
"--ploidy", str(dd.get_ploidy(data)),
"-o", tx_call_file, out["cns"]]
if gender:
cmd += ["--gender", gender]
if gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit call ploidy")
calls = {}
for outformat in ["bed", "vcf"]:
out_file = "%s.%s" % (os.path.splitext(call_file)[0], outformat)
calls[outformat] = out_file
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cmd = [os.path.join(os.path.dirname(sys.executable), "cnvkit.py"), "export",
outformat, "--sample-id", dd.get_sample_name(data),
"--ploidy", str(dd.get_ploidy(data)),
"-o", tx_out_file, call_file]
if gender and gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit export %s" % outformat)
out["call_file"] = call_file
out["vrn_bed"] = annotate.add_genes(calls["bed"], data)
effects_vcf, _ = effects.add_to_vcf(calls["vcf"], data, "snpeff")
out["vrn_file"] = effects_vcf or calls["vcf"]
return out
def _add_bed_to_output(out, data):
"""Call ploidy and convert into BED representation.
"""
call_file = "%s-call%s" % os.path.splitext(out["cns"])
gender = dd.get_gender(data)
if not utils.file_exists(call_file):
with file_transaction(data, call_file) as tx_call_file:
cmd = [os.path.join(os.path.dirname(sys.executable), "cnvkit.py"), "call",
"--ploidy", str(dd.get_ploidy(data)),
"-o", tx_call_file, out["cns"]]
if gender:
cmd += ["--gender", gender]
if gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit call ploidy")
out_file = "%s.bed" % os.path.splitext(call_file)[0]
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cmd = [os.path.join(os.path.dirname(sys.executable), "cnvkit.py"), "export",
"bed", "--sample-id", dd.get_sample_name(data),
"--ploidy", str(dd.get_ploidy(data)),
"-o", tx_out_file, call_file]
if gender and gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit export BED")
out["call_file"] = call_file
out["vrn_file"] = annotate.add_genes(out_file, data)
return out
def _check_copy_number_changes(svtype, cn, minor_cn, data):
"""Check if copy number changes match the expected svtype.
"""
if svtype == "LOH" and minor_cn == 0:
return svtype
elif svtype == "amplification" and cn > dd.get_ploidy(data):
return svtype
else:
return "std"
def _check_copy_number_changes(svtype, cn, minor_cn, data):
"""Check if copy number changes match the expected svtype.
"""
if svtype == "LOH" and minor_cn == 0:
return svtype
elif svtype == "amplification" and cn > dd.get_ploidy(data):
return svtype
else:
return "std"
def _add_variantcalls_to_output(out, data, is_somatic=False):
"""Call ploidy and convert into VCF and BED representations.
"""
call_file = "%s-call%s" % os.path.splitext(out["cns"])
gender = population.get_gender(data)
if not utils.file_exists(call_file):
with file_transaction(data, call_file) as tx_call_file:
cmd = [
os.path.join(os.path.dirname(sys.executable), "cnvkit.py"),
"call", "--ploidy",
str(dd.get_ploidy(data)), "-o", tx_call_file, out["cns"]
]
small_vrn_files = _compatible_small_variants(data)
if len(small_vrn_files) > 0 and _cna_has_values(out["cns"]):
cmd += ["-v", small_vrn_files[0]]
if not is_somatic:
cmd += ["-m", "clonal"]
if gender and gender.lower() != "unknown":
cmd += ["--gender", gender]
if gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit call ploidy")
calls = {}
for outformat in ["bed", "vcf"]:
out_file = "%s.%s" % (os.path.splitext(call_file)[0], outformat)
calls[outformat] = out_file
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cmd = [
os.path.join(os.path.dirname(sys.executable), "cnvkit.py"),
"export", outformat, "--sample-id",
dd.get_sample_name(data), "--ploidy",
str(dd.get_ploidy(data)), "-o", tx_out_file, call_file
]
if gender and gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit export %s" % outformat)
out["call_file"] = call_file
out["vrn_bed"] = annotate.add_genes(calls["bed"], data)
effects_vcf, _ = effects.add_to_vcf(calls["vcf"], data, "snpeff")
out["vrn_file"] = effects_vcf or calls["vcf"]
return out
def cnv_to_event(name, data):
"""Convert a CNV to an event name.
"""
ploidy = dd.get_ploidy(data)
if name.startswith("cnv"):
num = max([int(x) for x in name.split("_")[0].replace("cnv", "").split(";")])
if num < ploidy:
return "DEL"
elif num > ploidy:
return "DUP"
else:
return name
else:
return name
def cnv_to_event(name, data):
"""Convert a CNV to an event name.
"""
ploidy = dd.get_ploidy(data)
if name.startswith("cnv"):
num = max([int(x) for x in name.split("_")[0].replace("cnv", "").split(";")])
if num < ploidy:
return "DEL"
elif num > ploidy:
return "DUP"
else:
return name
else:
return name
def _configured_ploidy(items):
ploidies = collections.defaultdict(set)
for data in items:
ploidy = dd.get_ploidy(data)
if isinstance(ploidy, dict):
for k, v in ploidy.items():
ploidies[k].add(v)
else:
ploidies["default"].add(ploidy)
out = {}
for k, vs in ploidies.items():
assert len(vs) == 1, "Multiple ploidies set for group calling: %s %s" % (k, list(vs))
out[k] = vs.pop()
return out
def _add_variantcalls_to_output(out, data):
"""Call ploidy and convert into VCF and BED representations.
"""
call_file = "%s-call%s" % os.path.splitext(out["cns"])
gender = dd.get_gender(data)
if not utils.file_exists(call_file):
with file_transaction(data, call_file) as tx_call_file:
cmd = [
os.path.join(os.path.dirname(sys.executable), "cnvkit.py"),
"call", "--ploidy",
str(dd.get_ploidy(data)), "-o", tx_call_file, out["cns"]
]
if gender:
cmd += ["--gender", gender]
if gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit call ploidy")
calls = {}
for outformat in ["bed", "vcf"]:
out_file = "%s.%s" % (os.path.splitext(call_file)[0], outformat)
calls[outformat] = out_file
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cmd = [
os.path.join(os.path.dirname(sys.executable), "cnvkit.py"),
"export", outformat, "--sample-id",
dd.get_sample_name(data), "--ploidy",
str(dd.get_ploidy(data)), "-o", tx_out_file, call_file
]
if gender and gender.lower() == "male":
cmd += ["--male-reference"]
do.run(cmd, "CNVkit export %s" % outformat)
out["call_file"] = call_file
out["vcf_file"] = calls["vcf"]
out["vrn_file"] = annotate.add_genes(calls["bed"], data)
return out
def _add_bed_to_output(out, data):
"""Add FreeBayes cnvmap BED-like representation to the output.
"""
out_file = "%s.bed" % os.path.splitext(out["cns"])[0]
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cmd = [os.path.join(os.path.dirname(sys.executable), "cnvkit.py"), "export",
"freebayes", "--name", dd.get_sample_name(data),
"--ploidy", str(dd.get_ploidy(data)),
"-o", tx_out_file, out["cns"]]
gender = dd.get_gender(data)
if gender:
cmd += ["--gender", gender]
do.run(cmd, "CNVkit export FreeBayes BED cnvmap")
out["vrn_file"] = out_file
return out
def _add_bed_to_output(out, data):
"""Add FreeBayes cnvmap BED-like representation to the output.
"""
out_file = "%s.bed" % os.path.splitext(out["cns"])[0]
if not utils.file_exists(out_file):
with file_transaction(data, out_file) as tx_out_file:
cmd = [os.path.join(os.path.dirname(sys.executable), "cnvkit.py"), "export",
"freebayes", "--name", dd.get_sample_name(data),
"--ploidy", str(dd.get_ploidy(data)),
"-o", tx_out_file, out["cns"]]
gender = dd.get_gender(data)
if gender:
cmd += ["--gender", gender]
do.run(cmd, "CNVkit export FreeBayes BED cnvmap")
out["vrn_file"] = out_file
return out
