def __init__(self, args):
self.indels = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_indel)
self.snvs = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_snv)
self.info = []
self.fmt = []
self.flt = []
self.contigs = []
self.other = []
self.chrom_order = []
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 run(args):
'''Main wrapper function for filtering MuTect VCF files'''
logger.info('<FILTER>LowDPN=Normal depth < {0.min_normal_depth}'.format(args))
logger.info('<FILTER>LowDPT=Tumor depth < {0.min_tumor_depth}'.format(args))
logger.info('<FILTER>TAF=(SAMPLE == TUMOR) && (TYPE==SOM) && AF < {0.min_alt_freq_tumor:.2f}'.format(args))
logger.info('<FILTER>NAF=(SAMPLE == NORMAL) && (TYPE==SOM) && AF >= {0.max_alt_freq_normal:.2f}'.format(args))
logger.info('<FILTER>LowBQ=BQ < {0.min_base_quality}'.format(args))
# New filter and info lines to add to the output vcf
filters = [
'##FILTER=<ID=LowDPN,Description="Normal depth < {0.min_normal_depth}">'.format(args),
'##FILTER=<ID=LowDPT,Description="Tumor depth < {0.min_tumor_depth}">'.format(args),
'##FILTER=<ID=TAF,Description="Tumor: (TYPE == SOM) && AF < {0.min_alt_freq_tumor:.2f}">'.format(args),
'##FILTER=<ID=NAF,Description="Normal: (TYPE == SOM) && AF >= {0.max_alt_freq_normal:.2f}">'.format(args),
'##FILTER=<ID=LowBQ,Description="BQ < {0.min_base_quality}">'.format(args),
'##FILTER=<ID=UK,Description="SS==5">'
]
info = [
'##INFO=<ID=NTYPE,Number=1,Type=String,Description="Normal type, can be WT, GERM, SOMATIC, LOH, PTM, UK">',
'##INFO=<ID=TTYPE,Number=1,Type=String,Description="Tumor type, can be WT, GERM, SOMATIC, LOH, PTM, UK">'
]
# Process the file
with open(args.output_vcf, 'wb') as o:
mutect_reader = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_vcf)
mutect_reader.Open()
mutect_reader.get_header()
mutect_reader.write_new_header(o, filters=filters, info=info)
mutect_reader.apply_filters(o, args, MutectRecord)
mutect_reader.Close()
def run(args):
'''Main wrapper function for filtering MuTect VCF files'''
logger.info(
'<FILTER>LowDPN=Normal depth < {0.min_normal_depth}'.format(args))
logger.info(
'<FILTER>LowDPT=Tumor depth < {0.min_tumor_depth}'.format(args))
logger.info(
'<FILTER>TAF=(SAMPLE == TUMOR) && (TYPE==SOM) && AF < {0.min_alt_freq_tumor:.2f}'
.format(args))
logger.info(
'<FILTER>NAF=(SAMPLE == NORMAL) && (TYPE==SOM) && AF >= {0.max_alt_freq_normal:.2f}'
.format(args))
logger.info('<FILTER>LowBQ=BQ < {0.min_base_quality}'.format(args))
# New filter and info lines to add to the output vcf
filters = [
'##FILTER=<ID=LowDPN,Description="Normal depth < {0.min_normal_depth}">'
.format(args),
'##FILTER=<ID=LowDPT,Description="Tumor depth < {0.min_tumor_depth}">'.
format(args),
'##FILTER=<ID=TAF,Description="Tumor: (TYPE == SOM) && AF < {0.min_alt_freq_tumor:.2f}">'
.format(args),
'##FILTER=<ID=NAF,Description="Normal: (TYPE == SOM) && AF >= {0.max_alt_freq_normal:.2f}">'
.format(args),
'##FILTER=<ID=LowBQ,Description="BQ < {0.min_base_quality}">'.format(
args), '##FILTER=<ID=UK,Description="SS==5">'
]
info = [
'##INFO=<ID=NTYPE,Number=1,Type=String,Description="Normal type, can be WT, GERM, SOMATIC, LOH, PTM, UK">',
'##INFO=<ID=TTYPE,Number=1,Type=String,Description="Tumor type, can be WT, GERM, SOMATIC, LOH, PTM, UK">'
]
# Process the file
with open(args.output_vcf, 'wb') as o:
mutect_reader = VcfReader(args.choice, args.tumor_name,
args.normal_name, args.input_vcf)
mutect_reader.Open()
mutect_reader.get_header()
mutect_reader.write_new_header(o, filters=filters, info=info)
mutect_reader.apply_filters(o, args, MutectRecord)
mutect_reader.Close()
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()
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
class StrelkaReader(object):
'''Object that allows for the simultaneous processing of SNV and InDel VCFs'''
pattern = re.compile('##contig=<ID=(.+),.+>')
def __init__(self, args):
self.indels = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_indel)
self.snvs = VcfReader(args.choice, args.tumor_name, args.normal_name, args.input_all_snv)
self.info = []
self.fmt = []
self.flt = []
self.contigs = []
self.other = []
self.chrom_order = []
def Open(self):
'''Opens both indel and snv files'''
self.indels.Open()
self.snvs.Open()
def Close(self):
'''Closes both indel and snv files'''
self.indels.Close()
self.snvs.Close()
def get_headers(self):
'''Loads both headers'''
self.indels.get_header()
self.snvs.get_header()
# Combine info
dic = {}
for i in self.indels.header.info + self.snvs.header.info:
k = i.split(',')[0]
dic[k] = i
[self.info.append(dic[i]) for i in dic]
# Combine format
dic = {}
for i in self.indels.header.fmt + self.snvs.header.fmt:
k = i.split(',')[0]
dic[k] = i
[self.fmt.append(dic[i]) for i in dic]
# Combine filter
dic = {}
for i in self.indels.header.flt + self.snvs.header.flt:
k = i.split(',')[0]
dic[k] = i
[self.flt.append(dic[i]) for i in dic]
# Get contigs
self.contigs = self.indels.header.contig
# Get other info
self.other = list(set(self.indels.header.other + self.snvs.header.other))
def write_new_header(self, o, filters, info):
# Write fmt, date, reference, and other
if self.indels.header.vcffmt: o.write(self.indels.header.vcffmt + '\n')
if self.indels.header.vcfdate: o.write(self.indels.header.vcfdate + '\n')
if self.other: o.write('\n'.join(self.other) + '\n')
if self.indels.header.ref: o.write(self.indels.header.ref + '\n')
# Write contigs
if self.contigs: o.write('\n'.join(self.contigs) + '\n')
# Write
# Write old and new info
if self.info: o.write('\n'.join(self.info) + '\n')
if info: o.write('\n'.join(info) + '\n')
# Write old and new filters
if self.flt: o.write('\n'.join(self.flt) + '\n')
if filters: o.write('\n'.join(filters) + '\n')
# Write formats
if self.fmt: o.write('\n'.join(self.fmt) + '\n')
# Write new header
new_header = "\t".join(self.snvs.header.header[:9]) + '\t' + self.snvs.nname + '\t' + self.snvs.tname
o.write(new_header + '\n')
def apply_filters(self, args, o):
self.__get_chromosomes()
# First, load indels into a dic
dic = {}
for line in self.indels.fh:
indel_record = StrelkaIndelRecord(line.rstrip().split('\t'),
normal_idx=self.indels.nidx, tumor_idx=self.indels.tidx)
indel_record.apply_filters(args)
if indel_record.chrom not in dic: dic[indel_record.chrom] = {}
dic[indel_record.chrom][indel_record.pos] = indel_record
# Next, load snvs into the dic
for line in self.snvs.fh:
snv_record = StrelkaSnvRecord(line.rstrip().split('\t'),
normal_idx=self.snvs.nidx, tumor_idx=self.snvs.tidx)
snv_record.apply_filters(args)
if snv_record.chrom not in dic: dic[snv_record.chrom] = {}
dic[snv_record.chrom][snv_record.pos] = snv_record
# Now, write out the new VCF file with the correct order
for c in self.chrom_order:
if c in dic:
for p in sorted(dic[c]):
dic[c][p].write_record(o)
def __get_chromosomes(self):
'''Parse out the chromosome order from the contig lines'''
for i in self.contigs:
self.chrom_order.append(self.pattern.match(i).groups()[0])
class VirmidReader(object):
'''Object that allows for the simultaneous processing of germ, loh, and som VCFs'''
pattern = re.compile('##contig=<ID=(.+),.+>')
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 Open(self):
'''Opens germline, loh, and somatic files'''
self.germline.Open()
self.loh.Open()
self.somatic.Open()
def Close(self):
'''Closes germline, loh, and somatic files'''
self.germline.Close()
self.loh.Close()
self.somatic.Close()
def get_headers(self):
'''Loads all headers'''
self.germline.get_header()
self.loh.get_header()
self.somatic.get_header()
# Combine info
dic = {}
for i in self.germline.header.info + self.loh.header.info + self.somatic.header.info:
k = i.split(',')[0]
dic[k] = i
[self.info.append(dic[i]) for i in dic]
# Combine filter
dic = {}
for i in self.germline.header.flt + self.loh.header.flt + self.somatic.header.flt:
k = i.split(',')[0]
dic[k] = i
[self.flt.append(dic[i]) for i in dic]
# Get other info
self.other = list(set(self.germline.header.other + self.loh.header.other + self.somatic.header.other))
self.other.append('##normalSampleName={0.normal_name}'.format(self))
self.other.append('##tumorSampleName={0.tumor_name}'.format(self))
def write_new_header(self, o, filters, info):
'''Writes the new header lines to the output VCF'''
# Write fmt, date, reference, and other
if self.germline.header.vcffmt: o.write(self.germline.header.vcffmt + '\n')
if self.germline.header.vcfdate: o.write(self.germline.header.vcfdate + '\n')
if self.other: o.write('\n'.join(self.other) + '\n')
if self.germline.header.src: o.write(self.germline.header.src + '\n')
if self.germline.header.ref: o.write(self.germline.header.ref + '\n')
# Write contigs
if self.contigs: o.write('\n'.join(self.contigs) + '\n')
# Write old and new info
if self.info: o.write('\n'.join(self.info) + '\n')
if info: o.write('\n'.join(info) + '\n')
# Write old and new filters
if self.flt: o.write('\n'.join(self.flt) + '\n')
if filters: o.write('\n'.join(filters) + '\n')
# Write new header
new_header = "\t".join(self.somatic.header.header[:9])
o.write(new_header + '\n')
def apply_filters(self, args, o):
'''Wrapper function for applying filters to germline, LOH, and somatic VCFs'''
self.__get_chromosomes()
# First, load germline into a dic
dic = {}
for line in self.germline.fh:
germline_record = VirmidGermlineRecord(line.rstrip().split('\t'))
germline_record.apply_filters(args)
if germline_record.chrom not in dic: dic[germline_record.chrom] = {}
dic[germline_record.chrom][germline_record.pos] = germline_record
# Next, load LOH into the dic
for line in self.loh.fh:
loh_record = VirmidLohRecord(line.rstrip().split('\t'))
loh_record.apply_filters(args)
if loh_record.chrom not in dic: dic[loh_record.chrom] = {}
dic[loh_record.chrom][loh_record.pos] = loh_record
# Finally, load somatic into the dic
for line in self.somatic.fh:
somatic_record = VirmidSomaticRecord(line.rstrip().split('\t'))
somatic_record.apply_filters(args)
if somatic_record.chrom not in dic: dic[somatic_record.chrom] = {}
dic[somatic_record.chrom][somatic_record.pos] = somatic_record
# Now, write out the new VCF file with the correct order
for c in self.chrom_order:
if c in dic:
for p in sorted(dic[c]):
dic[c][p].write_record(o)
def __get_chromosomes(self):
'''Parse out the chromosome order from the contig lines'''
for i in self.contigs:
self.chrom_order.append(self.pattern.match(i).groups()[0])