def _add_reject_flag(in_file, config):
"""Add REJECT flag to all records that aren't flagged somatic
(SS=2)"""
Filter = namedtuple('Filter', ['id', 'desc'])
reject_filter = Filter(id='REJECT',
desc='Rejected as non-SOMATIC or by quality')
# NOTE: PyVCF will write an uncompressed VCF
base, ext = utils.splitext_plus(in_file)
name = "rejectfix"
out_file = "{0}-{1}{2}".format(base, name, ".vcf")
if utils.file_exists(in_file):
reader = vcf.VCFReader(filename=in_file)
# Add info to the header of the reader
reader.filters["REJECT"] = reject_filter
with file_transaction(out_file) as tx_out_file:
with open(tx_out_file, "wb") as handle:
writer = vcf.VCFWriter(handle, template=reader)
for record in reader:
if "SS" in record.INFO:
# VarScan encodes it as a string
# TODO: Set it as integer when cleaning
if record.INFO["SS"] != "2":
record.add_filter("REJECT")
writer.write_record(record)
# Re-compress the file
out_file = bgzip_and_index(out_file, config)
move_vcf(in_file, "{0}.orig".format(in_file))
move_vcf(out_file, in_file)
with open(out_file, "w") as out_handle:
out_handle.write("Moved to {0}".format(in_file))
def _add_reject_flag(in_file, config):
"""Add REJECT flag to all records that aren't flagged somatic
(SS=2)"""
Filter = namedtuple('Filter', ['id', 'desc'])
reject_filter = Filter(id='REJECT',
desc='Rejected as non-SOMATIC or by quality')
# NOTE: PyVCF will write an uncompressed VCF
base, ext = utils.splitext_plus(in_file)
name = "rejectfix"
out_file = "{0}-{1}{2}".format(base, name, ".vcf")
if utils.file_exists(in_file):
reader = vcf.VCFReader(filename=in_file)
# Add info to the header of the reader
reader.filters["REJECT"] = reject_filter
with file_transaction(config, out_file) as tx_out_file:
with open(tx_out_file, "wb") as handle:
writer = vcf.VCFWriter(handle, template=reader)
for record in reader:
if "SS" in record.INFO:
# VarScan encodes it as a string
# TODO: Set it as integer when cleaning
if record.INFO["SS"] != "2":
record.add_filter("REJECT")
writer.write_record(record)
# Re-compress the file
out_file = bgzip_and_index(out_file, config)
move_vcf(in_file, "{0}.orig".format(in_file))
move_vcf(out_file, in_file)
with open(out_file, "w") as out_handle:
out_handle.write("Moved to {0}".format(in_file))
def fix_somatic_calls(in_file, config):
"""Fix somatic variant output, standardize it to the SOMATIC flag.
"""
if vcf is None:
raise ImportError("Require PyVCF for manipulating cancer VCFs")
# HACK: Needed to replicate the structure used by PyVCF
Info = namedtuple('Info', ['id', 'num', 'type', 'desc'])
somatic_info = Info(id='SOMATIC', num=0, type='Flag', desc='Somatic event')
Filter = namedtuple('Filter', ['id', 'desc'])
reject_filter = Filter(id='REJECT',
desc='Rejected as non-SOMATIC or by quality')
# NOTE: PyVCF will write an uncompressed VCF
base, ext = utils.splitext_plus(in_file)
name = "somaticfix"
out_file = "{0}-{1}{2}".format(base, name, ".vcf")
if utils.file_exists(in_file):
reader = vcf.VCFReader(filename=in_file)
# Add info to the header of the reader
reader.infos["SOMATIC"] = somatic_info
reader.filters["REJECT"] = reject_filter
for ext in [".gz", ".gz.tbi"]:
if os.path.exists(out_file + ext):
os.remove(out_file + ext)
with file_transaction(config, out_file) as tx_out_file:
with open(tx_out_file, "wb") as handle:
writer = vcf.VCFWriter(handle, template=reader)
for record in reader:
# Handle FreeBayes
is_somatic = False
if "VT" in record.INFO:
if record.INFO["VT"] == "somatic":
record.add_info("SOMATIC", True)
is_somatic = True
# Discard old record
del record.INFO["VT"]
if not is_somatic:
record.add_filter("REJECT")
writer.write_record(record)
# Re-compress the file
out_file = bgzip_and_index(out_file, config)
move_vcf(in_file, "{0}.orig".format(in_file))
move_vcf(out_file, in_file)
with open(out_file, "w") as out_handle:
out_handle.write("Moved to {0}".format(in_file))
def fix_somatic_calls(in_file, config):
"""Fix somatic variant output, standardize it to the SOMATIC flag.
"""
if vcf is None:
raise ImportError("Require PyVCF for manipulating cancer VCFs")
# HACK: Needed to replicate the structure used by PyVCF
Info = namedtuple('Info', ['id', 'num', 'type', 'desc'])
somatic_info = Info(id='SOMATIC', num=0, type='Flag', desc='Somatic event')
Filter = namedtuple('Filter', ['id', 'desc'])
reject_filter = Filter(id='REJECT', desc='Rejected as non-SOMATIC or by quality')
# NOTE: PyVCF will write an uncompressed VCF
base, ext = utils.splitext_plus(in_file)
name = "somaticfix"
out_file = "{0}-{1}{2}".format(base, name, ".vcf")
if utils.file_exists(in_file):
reader = vcf.VCFReader(filename=in_file)
# Add info to the header of the reader
reader.infos["SOMATIC"] = somatic_info
reader.filters["REJECT"] = reject_filter
for ext in [".gz", ".gz.tbi"]:
if os.path.exists(out_file + ext):
os.remove(out_file + ext)
with file_transaction(config, out_file) as tx_out_file:
with open(tx_out_file, "wb") as handle:
writer = vcf.VCFWriter(handle, template=reader)
for record in reader:
# Handle FreeBayes
is_somatic = False
if "VT" in record.INFO:
if record.INFO["VT"] == "somatic":
record.add_info("SOMATIC", True)
is_somatic = True
# Discard old record
del record.INFO["VT"]
if not is_somatic:
record.add_filter("REJECT")
writer.write_record(record)
# Re-compress the file
out_file = bgzip_and_index(out_file, config)
move_vcf(in_file, "{0}.orig".format(in_file))
move_vcf(out_file, in_file)
with open(out_file, "w") as out_handle:
out_handle.write("Moved to {0}".format(in_file))
def clean_vcf_output(orig_file, clean_fn, config, name="clean"):
"""Provide framework to clean a file in-place, with the specified clean
function.
"""
base, ext = utils.splitext_plus(orig_file)
out_file = "{0}-{1}{2}".format(base, name, ext)
if not utils.file_exists(out_file):
with open(orig_file) as in_handle:
with file_transaction(config, out_file) as tx_out_file:
with open(tx_out_file, "w") as out_handle:
for line in in_handle:
update_line = clean_fn(line)
if update_line:
out_handle.write(update_line)
move_vcf(orig_file, "{0}.orig".format(orig_file))
move_vcf(out_file, orig_file)
with open(out_file, "w") as out_handle:
out_handle.write("Moved to {0}".format(orig_file))
def clean_vcf_output(orig_file, clean_fn, config, name="clean"):
"""Provide framework to clean a file in-place, with the specified clean
function.
"""
base, ext = utils.splitext_plus(orig_file)
out_file = "{0}-{1}{2}".format(base, name, ext)
if not utils.file_exists(out_file):
with open(orig_file) as in_handle:
with file_transaction(config, out_file) as tx_out_file:
with open(tx_out_file, "w") as out_handle:
for line in in_handle:
update_line = clean_fn(line)
if update_line:
out_handle.write(update_line)
move_vcf(orig_file, "{0}.orig".format(orig_file))
move_vcf(out_file, orig_file)
with open(out_file, "w") as out_handle:
out_handle.write("Moved to {0}".format(orig_file))