def run(args):
'''Main wrapper function for filtering VarScan2 files'''
# Print the filters to the log
logger.info('<FILTER>LowDPN=Normal DP < {0.min_normal_depth}'.format(args))
logger.info('<FILTER>LowDPT=Tumor DP < {0.min_tumor_depth}'.format(args))
logger.info('<FILTER>NAF=Normal FREQ >= {0.max_alt_freq_normal:.2f}'.format(args))
logger.info('<FILTER>TAF=Tumor FREQ < {0.min_alt_freq_tumor:.2f}'.format(args))
logger.info('<FILTER>SPV=SPV >= {0.pval_cutoff:.4e}'.format(args))
logger.info('<FILTER>GPV=GPV >= {0.pval_cutoff:.4e}'.format(args))
# Chromosome order list
contigs = load_contigs(args)
chrm_order = []
for i in contigs:
chrm_order.append(pattern.match(i).groups()[0])
## New filter and info lines
new_header = """##fileformat=VCFv4.1
##fileDate={fdate}
##source=VarScan2
##reference=file://{ref}
{contigs}
##INFO=<ID=SS,Number=1,Type=String,Description="Somatic status call SOMATIC=somatic, LOH=loss of het, GERM=germline">
##INFO=<ID=GPV,Number=1,Type=Float,Description="Variant p-value for germline events">
##INFO=<ID=SPV,Number=1,Type=Float,Description="Somatic p-value for Somatic/LOH events">
##INFO=<ID=INDEL,Number=0,Type=Flag,Description="If this is an InDel">
##FORMAT=<ID=RD,Number=1,Type=Integer,Description="Reads supporting reference in sample">
##FORMAT=<ID=AD,Number=1,Type=Integer,Description="Reads supporting variant in sample">
##FORMAT=<ID=FREQ,Number=1,Type=Float,Description="Variant allele frequency in sample">
##FORMAT=<ID=GT,Number=1,Type=String,Description="Concensus genotype call">
##FORMAT=<ID=DP,Number=4,Type=Integer,Description="REF reads +,REF read -, ALT reads +, ALT reads -">
##FILTER=<ID=LowDPN,Description="Normal DP < {0.min_normal_depth}">
##FILTER=<ID=LowDPT,Description="Tumor DP < {0.min_tumor_depth}">
##FILTER=<ID=NAF,Description="(SS==SOM) && (Normal FREQ >= {0.max_alt_freq_normal:.2f})">
##FILTER=<ID=TAF,Description="(SS==SOM) && (Tumor FREQ < {0.min_alt_freq_tumor:.2f})">
##FILTER=<ID=SPV,Description="(SS==SOM) && (SPV >= {0.pval_cutoff:.4e}">
##FILTER=<ID=GPV,Description="(SS==GERM) && (GPV >= {0.pval_cutoff:.4e}">
""".format(args,fdate=datetime.date.today(),ref=os.path.abspath(args.reference),
contigs="\n".join(contigs))
# CHROM header line
chromline = '#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\t{0}\t{1}\n'.format(
args.normal_name, args.tumor_name)
# Load InDels and SNVs
var_dict = {}
var_dict = load_indel(args, var_dict)
var_dict = load_snv(args, var_dict)
# Write the new VCF file
with open(args.output_vcf, 'wb') as o:
# Write header
o.write(new_header)
# Write CHROM line
o.write(chromline)
# Write results
for c in chrm_order:
if c in var_dict:
for p in sorted(var_dict[c]):
o.write(var_dict[c][p] + '\n')
def run(args):
'''Main wrapper function for filtering SomaticSniper VCF files'''
# Print the filters to the log
logger.info(
'<FILTER>LowDPN="Normal DP < {0.min_normal_depth}"'.format(args))
logger.info('<FILTER>LowDPT="Tumor DP < {0.min_tumor_depth}"'.format(args))
logger.info('<FILTER>LowMQT="Tumor MQ < {0.min_mapq_tumor}"'.format(args))
logger.info(
'<FILTER>LowMQN="Normal MQ < {0.min_mapq_normal}"'.format(args))
logger.info('<FILTER>LowGQT="Tumor GQ < {0.min_gq_tumor}"'.format(args))
logger.info('<FILTER>LowGQN="Normal GQ < {0.min_gq_normal}"'.format(args))
logger.info(
'<FILTER>LowScore="Somatic score < {0.min_somatic_score}"'.format(
args))
# New info and filter lines
info = [
'##INFO=<ID=NTYPE,Number=1,Type=String,Description="Normal type, can be REF,GERM,SOMATIC,LOH,UK">',
'##INFO=<ID=TTYPE,Number=1,Type=String,Description="Tumor type REF,GERM,SOMATIC,LOH,UK">'
]
filters = [
'##FILTER=<ID=PASS,Description="Accept as a confident somatic mutation">',
'##FILTER=<ID=LowDPN,Description="Normal DP < {0.min_normal_depth}">'.
format(args),
'##FILTER=<ID=LowDPT,Description="Tumor DP < {0.min_tumor_depth}">'.
format(args),
'##FILTER=<ID=LowMQT,Description="Tumor MQ < {0.min_mapq_tumor}">'.
format(args),
'##FILTER=<ID=LowMQN,Description="Normal MQ < {0.min_mapq_normal}">'.
format(args),
'##FILTER=<ID=LowGQT,Description="Tumor GQ < {0.min_gq_tumor}">'.
format(args),
'##FILTER=<ID=LowGQN,Description="Normal GQ < {0.min_gq_normal}">'.
format(args),
'##FILTER=<ID=LowScore,Description="Somatic score < {0.min_somatic_score}">'
.format(args), '##FILTER=<ID=UK,Description="Unknown variant type">'
]
# Load the contigs
contigs = load_contigs(args)
# Process the file
with open(args.output_vcf, 'wb') as o:
sniper_reader = VcfReader(args.choice, args.tumor_name,
args.normal_name, args.input_vcf)
sniper_reader.Open()
sniper_reader.get_header()
sniper_reader.write_new_header(o,
filters=filters,
info=info,
contigs=contigs)
sniper_reader.apply_filters(o, args, SniperRecord)
sniper_reader.Close()
logger.info('Filtered and formatted VCF file: {0}'.format(
os.path.abspath(args.output_vcf)))
def __init__(self, args):
self.germline = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_germline)
self.loh = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_loh)
self.somatic = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_somatic)
self.info = []
self.flt = []
self.contigs = load_contigs(args)
self.other = []
self.chrom_order = []
self.tumor_name = args.tumor_name
self.normal_name = args.normal_name
def run(args):
'''Main wrapper function for filtering Shimmer VCF files'''
# Print the filters to the log
logger.info('<FILTER>LowDPN=Normal DP < {0.min_normal_depth}'.format(args))
logger.info('<FILTER>LowDPT=Tumor DP < {0.min_tumor_depth}'.format(args))
logger.info('<FILTER>TAF=Tumor AF < {0.min_alt_freq_tumor:.3f}>'.format(args))
logger.info('<FILTER>NAF=Normal AF >= {0.max_alt_freq_normal:.3f}>'.format(args))
logger.info('<FILTER>LowQual=QUAL < {0.min_qual}>'.format(args))
# New filter and format lines
formats = ['##FORMAT=<ID=AF,Number=1,Type=Float,Description="Ratio of reads with alternate base">']
filters = [
'##FILTER=<ID=PASS,Description="Accept as a confident somatic mutation">',
'##FILTER=<ID=LowDPN,Description="Normal DP < {0.min_normal_depth}">'.format(args),
'##FILTER=<ID=LowDPT,Description="Tumor DP < {0.min_tumor_depth}">'.format(args),
'##FILTER=<ID=TAF,Description="Tumor AF < {0.min_alt_freq_tumor:.3f}">'.format(args),
'##FILTER=<ID=NAF,Description="Normal AF >= {0.max_alt_freq_normal:.3f}">'.format(args),
'##FILTER=<ID=LowQual,Description="QUAL < {0.min_qual}">'.format(args)
]
# Load the VarSifter dictionary
varsifter = load_varsifter(args)
# Load the contigs
reffile = '##reference=file://' + os.path.abspath(args.reference)
contigs = load_contigs(args)
# Process the file
with open(args.output_vcf, 'wb') as o:
shimmer_reader = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_vcf)
shimmer_reader.Open()
shimmer_reader.get_header()
shimmer_reader.write_new_header(o, filters=filters, formats=formats,
refpath=reffile, contigs=contigs)
shimmer_reader.apply_filters(o, args, ShimmerRecord, vsdict=varsifter)
shimmer_reader.Close()