def run(self):
# Phase 1 - Detection of BarCode
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting BarCode Analysis \n" % (self.get_time(),))
self.bc.simple_approach()
sys.stderr.write("[%s] Analyzed BarCodes \n" % (self.get_time(),))
self.bc.write_barcodes(self.barcodes)
sys.stderr.write("[%s] Wrote BarCodes\n" % (self.get_time(),))
# Phase 2 - Rewrite BAM
sys.stderr.write("[%s] Starting Sort and Rewrite BAM\n" % (self.get_time(),))
self.bc.load_barcodes(self.barcodes)
sys.stderr.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
self.bc.bam.reset()
self.bc.sort_and_rewrite_bam(self.rewritten_bam)
pysam.sort("-n", self.rewritten_bam, self.rewritten_sorted_bam.replace(".bam", ""))
sys.stderr.write("[%s] Sort and Rewrite BAM\n" % (self.get_time(),))
# Phase 3 - Build Consensus
self.consensus = Consensus(self.rewritten_sorted_bam, self.ref)
sys.stderr.write("[%s] Starting Consensus Building\n" % (self.get_time(),))
self.consensus.build()
sys.stderr.write("[%s] Built and Calculated Consensus\n" % (self.get_time(),))
self.consensus.infer_consensus(self.consensus_reference)
sys.stderr.write("[%s] Inferred Consensus\n" % (self.get_time(),))
# Phase 4 - Call Variants and Haplotypes
self.consensus.output_consensus_genomes(self.consensus_genomes)
sys.stderr.write("[%s] Output Consensus Genomes\n" % (self.get_time(),))
self.consensus.output_haplotype_distribution(self.haplotype_distribution)
sys.stderr.write("[%s] Output Haplotype Distribution\n" % (self.get_time(),))
self.ovcf = VCF(self.vcf, crossmap=self.crossmap)
self.ovcf.get_variants(self.ref.sequence,
self.consensus.consensus_genomes)
self.ovcf.output_vcf(self.ref.sequence)
sys.stderr.write("[%s] Output VCF\n" % (self.get_time(),))
# Phase 5 - Summary Statistics and Chain Files
f_out = open(self.out, "w")
self.consensus.output_consensus_coverage(f_out)
self.ovcf.output_variants_distribution(f_out)
self.bc.output_reads_in_barcode_distribution(f_out)
f_out.close()
sys.stderr.write("[%s] Output Summary Statistics\n" % (self.get_time(),))
self.ochain = Chain(self.chain)
self.ochain.output_chain(self.ref,
self.consensus.inferred_consensus,
self.consensus.inferred_structure)
sys.stderr.write("[%s] Output Chain File\n" % (self.get_time(),))
def as_indel(self, ref_fasta):
chrom = self.chroms[0].lstrip('chr')
pos = self.breaks[0]
ref = alt = None
size = self.get_size()
if self.rearrangement == 'del':
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[1] - 1).upper()
alt = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[0]).upper()
else:
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[1]).upper()
alt = ref + self.novel_seq.upper()
id = self.id
qual = '.'
filter = '.'
info = {
'BKPTID': ','.join(self.contigs),
}
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
# repeat contraction
if self.rearrangement == 'del' and self.repeat_seq is not None:
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
if ref is not None and alt is not None:
fields = [
chrom, pos, id, ref, alt, qual, filter,
VCF.info_dict_to_str(info)
]
return '\t'.join(map(str, fields))
def main(): input = ComLine(sys.argv[1:]) vcf_file = VCF(input.args.vcf, input.args.thin, input.args.maf, input.args.indcov, input.args.snpcov, input.args.bi) #if input.args.filter == True: # vcf_file.convert_filter() #else: #convert to Plink vcf_file.convert() populations = Popmap(input.args.popmap) vcf_file.plink() vcf_file.print_populations(populations) admix_run = Admixture(vcf_file.prefix, input.args.np, input.args.minK, input.args.maxK, input.args.rep, input.args.cv) admix_run.admix() admix_run.create_zip() admix_run.loglik() admix_run.print_cv()
def as_indel(self, ref_fasta):
chrom = self.chroms[0].lstrip('chr')
pos = self.breaks[0]
ref = alt = None
size = self.get_size()
if self.rearrangement == 'del':
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[1] - 1).upper()
alt = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[0]).upper()
else:
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[1]).upper()
alt = ref + self.novel_seq.upper()
id = self.id
qual = '.'
filter = '.'
info = {
'BKPTID':','.join(self.contigs),
}
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
# repeat contraction
if self.rearrangement == 'del' and self.repeat_seq is not None:
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
if ref is not None and alt is not None:
fields = [chrom, pos, id, ref, alt, qual, filter, VCF.info_dict_to_str(info)]
return '\t'.join(map(str, fields))
def assemble_consensus_genomes(self):
# build consensus
self.consensus = Consensus(self.rewritten_sorted_bam, self.ref)
sys.stderr.write("[%s] Starting Consensus Building\n" % (self.get_time(),))
self.consensus.build(debug=self.debug)
sys.stderr.write("[%s] Built and Calculated Consensus\n" % (self.get_time(),))
self.consensus.infer_consensus(self.consensus_reference)
sys.stderr.write("[%s] Inferred Consensus\n" % (self.get_time(),))
self.consensus.output_consensus_genomes(self.consensus_genomes)
sys.stderr.write("[%s] Output Consensus Genomes\n" % (self.get_time(),))
self.consensus.output_haplotype_distribution(self.haplotype_distribution)
sys.stderr.write("[%s] Output Haplotype Distribution\n" % (self.get_time(),))
self.quark = Quark(self.ref.sequence)
self.quark.distance_matrix(sorted(self.consensus.freq_distribution.items(),
key=lambda q: q[1],
reverse=True))
self.quark.graph_it()
self.quark.rank_it(self.rank)
sys.stderr.write("[%s] Output Quark Analysis\n" % (self.get_time(),))
self.ovcf = VCF(self.vcf, crossmap=self.crossmap)
self.ovcf.get_variants(self.ref.sequence,
self.consensus.consensus_genomes)
self.ovcf.output_vcf(self.ref.sequence)
sys.stderr.write("[%s] Output VCF\n" % (self.get_time(),))
self.summary_statistics()
sys.stderr.write("[%s] Output Summary Statistics\n" % (self.get_time(),))
self.ochain = Chain(self.chain)
self.ochain.output_chain(self.ref,
self.consensus.inferred_consensus,
self.consensus.inferred_structure)
sys.stderr.write("[%s] Output Chain File\n" % (self.get_time(),))
import sys
import matplotlib.pyplot as plt
from matplotlib_venn import venn2
sys.path.append("/home/zhusitao/ngs-tools/format/")
from vcf import VCF
vcf1 = sys.argv[1]
title1 = sys.argv[2]
vcf2 = sys.argv[3]
title2 = sys.argv[4]
list1 = []
list2 = []
v1 = VCF(vcf1)
for record in v1.readVCF():
chrom, pos, ref, ale = record.CHROM, record.POS, record.REF, record.ALT
list1.append(chrom + pos + ref + ale)
v2 = VCF(vcf2)
for record in v2.readVCF():
chrom, pos, ref, ale = record.CHROM, record.POS, record.REF, record.ALT
list2.append(chrom + pos + ref + ale)
set1 = set(list1)
set2 = set(list2)
inter = set1 & set2
inter_len = len(inter)
set1_len = len(set1)
set2_len = len(set2)
def main():
input = ComLine(sys.argv[1:])
pops = Popmap(input.args.popmap)
vcf = VCF(input.args.vcf, pops)
vcf.printFile(input.args.out)
vcf.printPrivate(input.args.out)
def main():
# argument method
parser = argparse.ArgumentParser()
# positional argument
parser.add_argument('-d',
'--directory',
'--dir',
help='parent directory of the samples i.e. Sample_054',
required=True)
# optional argument
parser.add_argument(
'-f',
'--filter',
action='store_true',
help=
'use argument to keep the following columns: CHROM, POS, REF, ALT, genotype info. \n'
'Note: if option selection, the -d can either be the vcf.gz or the parent directory. '
'If vcf.gz, it will perform filtering on that file. Else, it will perform filtering on'
'all the files in the hierarchical directory.')
# optional argument
parser.add_argument(
'-m',
'--merge',
action='store_true',
help=
'use argument to merge all filtered.vcf.gz files in the parent directory'
)
# optional argument
parser.add_argument(
'-o',
'--output',
help='directory where to output the filtered or merged files')
# optional argument
parser.add_argument('-ht',
'--homozygous_test',
action='store_true',
help='use argument to collect homozygous statistics')
# optional argument
parser.add_argument(
'-s',
'--subset',
action='store_true',
help='use argument to subset the vcf file based on chromosomes')
# optional argument
parser.add_argument(
'-c',
'--chromosome',
help=
'use argument to select the chromosome number on which to subset on')
# optional argument
parser.add_argument(
'-n',
'--number_sites',
help='use argument to select the number of line on which to subset on',
type=int)
# optional argument
parser.add_argument('-p',
'--phase',
action='store_true',
help='use argument to select the phasing test')
parser.add_argument('-lc',
'--list_chromosomes',
action='store_true',
help='list all the chromosomes in a vcf.gz '
'file')
args = parser.parse_args()
working_directory, output_directory, chromosome, filter_merge = process_arguments(
c_directory=args.directory,
o_directory=args.output,
filter_flag=args.filter,
merge_flag=args.merge,
chrom=args.chromosome,
arg_parser=parser)
# create VCF object
vcf = VCF()
if args.list_chromosomes:
vcf.read_files(c_dir=working_directory, vcf_filtered_file=False)
vcf.list_chrom(output_dir=output_directory)
# filtering and merging have to read all the vcf files on the subdirectories of the families
elif filter_merge:
# read all the vcf files for that family
vcf.read_files(c_dir=working_directory)
# filter columns of the vcf files
if args.filter:
vcf.filter(output_dir=output_directory)
else:
vcf.merge(output_dir=output_directory)
# subset, homozygous or phasing tests have to read a vcf file in the parent directory
else:
# for the subset, the chromosome should not be given in the read_file function
if args.subset:
# chromosome if required if goal is to subset since the file will be subsetted on it
if not args.chromosome:
raise ValueError(
'Chromosome number must be provided in order to perform subset'
)
# read all the vcf files for that family
vcf.read_files(c_dir=working_directory, vcf_filtered_file=True)
vcf.subset(chrom=args.chromosome,
output_dir=output_directory,
n_sites=args.number_sites)
else:
# read all the vcf files for that family
vcf.read_files(c_dir=working_directory,
vcf_filtered_file=True,
chrom=chromosome)
if args.homozygous_test:
# collect homozygous statistics
vcf.tests(output_dir=output_directory,
chrom=chromosome,
homozygous_test=True)
if args.phase:
vcf.tmp_test(output_dir=output_directory, chrom=chromosome)
def main():
input = ComLine(sys.argv[1:])
vcf_file = VCF(input.args.vcf, input.args.thin, input.args.maf,
input.args.mac, input.args.indcov, input.args.snpcov,
input.args.bi, input.args.remove)
#convert to Plink
populations = Popmap(input.args.popmap)
vcf_file.compIndLists(populations)
vcf_file.convert()
vcf_file.plink()
vcf_file.print_populations(populations)
vcf_file.print_individuals(populations)
admix_run = Admixture(vcf_file.prefix, input.args.np, input.args.minK,
input.args.maxK, input.args.rep, input.args.cv)
admix_run.admix()
admix_run.create_zip()
admix_run.loglik()
admix_run.print_cv()
def as_sv(self, ref_fasta, id_ext=None, info_ext=None, chrom_ext=None, pos_ext=None):
chrom = self.chroms[0] if chrom_ext is None else chrom_ext
pos = self.breaks[0] if pos_ext is None else pos_ext
chrom = chrom.lstrip('chr')
alt = None
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[0]).upper()
sv_len = self.get_size()
end = None
if type(sv_len) is int:
end = pos + sv_len
if self.rearrangement == 'del':
alt = '<DEL>'
sv_type = 'DEL'
if type(sv_len) is int:
sv_len = -1 * sv_len
end = pos - sv_len
elif self.rearrangement == 'dup':
alt = '<DUP:TANDEM>'
sv_type = 'DUP'
elif self.rearrangement == 'inv':
alt = '<INV>'
sv_type = 'INV'
elif self.rearrangement == 'ins':
alt = '<INS>'
sv_type = 'INS'
end = pos
id = self.id if id_ext is None else id_ext
qual = '.'
filter = '.'
info = {'SVTYPE': sv_type,
'END': end,
'BKPTID':','.join(self.contigs),
}
if end is not None:
info['END'] = end
if type(sv_len) is int:
info['SVLEN'] = sv_len
if sv_type == 'DUP':
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not None:
info['FLANKING_PAIRS'] = self.support['flanking']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-':
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
#print 'gmap', contig_breaks
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
cipos = '0,%d' % homol_len
if cipos is not None:
info['CIPOS'] = cipos
info['CIPOS'] = cipos
if homol_len is not None:
info['HOMLEN'] = homol_len
info['HOMLEN'] = homol_len
if homol_seq is not None:
info['HOMSEQ'] = homol_seq
info['HOMSEQ'] = homol_seq
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if key == 'SVLEN' and value == 'NA':
continue
info[key] = value
if ref is not None and alt is not None:
fields = [chrom, pos, id, ref, alt, qual, filter, VCF.info_dict_to_str(info)]
return '\t'.join(map(str, fields))
class BaseSeq(Helper):
def __init__(self, bam, barcodes=None, out=None, ref=None,
rewritten_bam=None,
consensus_reference=None,
consensus_genomes=None,
haplotype_distribution=None,
vcf=None,
chain=None,
crossmap=None,
export=None,
rank=None,
debug=None):
self.bam = bam
self.barcodes = barcodes
self.out = out
self.ref = Reference(ref)
self.rewritten_bam = rewritten_bam
self.rewritten_sorted_bam = rewritten_bam.replace(".bam", ".sorted.bam") if rewritten_bam else None
self.consensus_reference = consensus_reference
self.consensus_genomes = consensus_genomes
self.haplotype_distribution = haplotype_distribution
self.vcf = vcf
self.chain = chain
self.crossmap = crossmap
self.export = export
self.rank = rank
self.debug = int(debug)
def get_barcodes(self):
# simple approach - align, take soft-clipped, and use the arbitrary 20 bases
# intermediate approach - use the seed and extend approach
out = open(self.out, "w")
self.bc = BarCode(self.bam)
self.bc.simple_approach()
for k, v in sorted(self.bc.barcode_to_read.items()):
q = sorted(v)
out.write("%s\t%s\n" % (k, ",".join(q)))
out.close()
def error_correction_barcodes(self):
# start analysis
self.bc = BarCode(self.bam)
sys.stdout.write("[%s] Starting Error Correction Analysis\n" % (self.get_time(),))
# load barcodes
self.bc.load_barcodes(self.barcodes)
sys.stdout.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
# cluster barcodes
self.bc.cluster_barcodes()
sys.stdout.write("[%s] Clustered BarCodes\n" % (self.get_time(),))
def filter_barcodes(self, barcode, export="fastq"):
list_of_ids = []
with open(self.barcodes, "r") as f:
for line in f:
data = line.strip("\r\n").split("\t")
if barcode == data[0]:
list_of_ids = data[1].split(",")
break
self.bc = BarCode(self.bam)
self.bc.filter_and_export(list_of_ids, self.out, export=export)
def sort_and_rewrite_bam(self):
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting Sort and Rewrite BAM\n" % (self.get_time(),))
self.bc.load_barcodes(self.barcodes)
sys.stderr.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
self.bc.sort_and_rewrite_bam(self.rewritten_bam)
pysam.sort("-n", self.rewritten_bam, self.rewritten_sorted_bam.replace(".bam", ""))
sys.stderr.write("[%s] Sort and Rewrite BAM\n" % (self.get_time(),))
def split_bam_by_barcode(self):
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting procedure to split BAM by barcode\n" % (self.get_time(),))
self.bc.split_bam_into_barcodes(self.ref, self.out, self.export)
sys.stderr.write("[%s] Finished splitting BAM by barcode id\n" % (self.get_time(),))
def assemble_consensus_genomes(self):
# build consensus
self.consensus = Consensus(self.rewritten_sorted_bam, self.ref)
sys.stderr.write("[%s] Starting Consensus Building\n" % (self.get_time(),))
self.consensus.build(debug=self.debug)
sys.stderr.write("[%s] Built and Calculated Consensus\n" % (self.get_time(),))
self.consensus.infer_consensus(self.consensus_reference)
sys.stderr.write("[%s] Inferred Consensus\n" % (self.get_time(),))
self.consensus.output_consensus_genomes(self.consensus_genomes)
sys.stderr.write("[%s] Output Consensus Genomes\n" % (self.get_time(),))
self.consensus.output_haplotype_distribution(self.haplotype_distribution)
sys.stderr.write("[%s] Output Haplotype Distribution\n" % (self.get_time(),))
self.quark = Quark(self.ref.sequence)
self.quark.distance_matrix(sorted(self.consensus.freq_distribution.items(),
key=lambda q: q[1],
reverse=True))
self.quark.graph_it()
self.quark.rank_it(self.rank)
sys.stderr.write("[%s] Output Quark Analysis\n" % (self.get_time(),))
self.ovcf = VCF(self.vcf, crossmap=self.crossmap)
self.ovcf.get_variants(self.ref.sequence,
self.consensus.consensus_genomes)
self.ovcf.output_vcf(self.ref.sequence)
sys.stderr.write("[%s] Output VCF\n" % (self.get_time(),))
self.summary_statistics()
sys.stderr.write("[%s] Output Summary Statistics\n" % (self.get_time(),))
self.ochain = Chain(self.chain)
self.ochain.output_chain(self.ref,
self.consensus.inferred_consensus,
self.consensus.inferred_structure)
sys.stderr.write("[%s] Output Chain File\n" % (self.get_time(),))
def summary_statistics(self):
# coverage per genome
# variants per genome
# estimate PCR and sequencing errors
# barcode distribution
f_out = open(self.out, "w")
self.bc = BarCode(self.bam) #TEMP
self.bc.load_barcodes(self.barcodes) #TEMP
self.consensus.output_consensus_coverage(f_out)
self.ovcf.output_variants_distribution(f_out)
self.bc.output_reads_in_barcode_distribution(f_out)
f_out.close()
def run(self):
# Phase 1 - Detection of BarCode
self.bc = BarCode(self.bam)
sys.stderr.write("[%s] Starting BarCode Analysis \n" % (self.get_time(),))
self.bc.simple_approach()
sys.stderr.write("[%s] Analyzed BarCodes \n" % (self.get_time(),))
self.bc.write_barcodes(self.barcodes)
sys.stderr.write("[%s] Wrote BarCodes\n" % (self.get_time(),))
# Phase 2 - Rewrite BAM
sys.stderr.write("[%s] Starting Sort and Rewrite BAM\n" % (self.get_time(),))
self.bc.load_barcodes(self.barcodes)
sys.stderr.write("[%s] Loaded BarCodes\n" % (self.get_time(),))
self.bc.bam.reset()
self.bc.sort_and_rewrite_bam(self.rewritten_bam)
pysam.sort("-n", self.rewritten_bam, self.rewritten_sorted_bam.replace(".bam", ""))
sys.stderr.write("[%s] Sort and Rewrite BAM\n" % (self.get_time(),))
# Phase 3 - Build Consensus
self.consensus = Consensus(self.rewritten_sorted_bam, self.ref)
sys.stderr.write("[%s] Starting Consensus Building\n" % (self.get_time(),))
self.consensus.build()
sys.stderr.write("[%s] Built and Calculated Consensus\n" % (self.get_time(),))
self.consensus.infer_consensus(self.consensus_reference)
sys.stderr.write("[%s] Inferred Consensus\n" % (self.get_time(),))
# Phase 4 - Call Variants and Haplotypes
self.consensus.output_consensus_genomes(self.consensus_genomes)
sys.stderr.write("[%s] Output Consensus Genomes\n" % (self.get_time(),))
self.consensus.output_haplotype_distribution(self.haplotype_distribution)
sys.stderr.write("[%s] Output Haplotype Distribution\n" % (self.get_time(),))
self.ovcf = VCF(self.vcf, crossmap=self.crossmap)
self.ovcf.get_variants(self.ref.sequence,
self.consensus.consensus_genomes)
self.ovcf.output_vcf(self.ref.sequence)
sys.stderr.write("[%s] Output VCF\n" % (self.get_time(),))
# Phase 5 - Summary Statistics and Chain Files
f_out = open(self.out, "w")
self.consensus.output_consensus_coverage(f_out)
self.ovcf.output_variants_distribution(f_out)
self.bc.output_reads_in_barcode_distribution(f_out)
f_out.close()
sys.stderr.write("[%s] Output Summary Statistics\n" % (self.get_time(),))
self.ochain = Chain(self.chain)
self.ochain.output_chain(self.ref,
self.consensus.inferred_consensus,
self.consensus.inferred_structure)
sys.stderr.write("[%s] Output Chain File\n" % (self.get_time(),))
def assemble_genomes(self):
pass
def assemble_genomes_from_fastq(self):
pass
def as_sv(self,
ref_fasta,
id_ext=None,
info_ext=None,
chrom_ext=None,
pos_ext=None):
chrom = self.chroms[0] if chrom_ext is None else chrom_ext
pos = self.breaks[0] if pos_ext is None else pos_ext
chrom = chrom.lstrip('chr')
alt = None
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[0]).upper()
sv_len = self.get_size()
end = None
if type(sv_len) is int:
end = pos + sv_len
if self.rearrangement == 'del':
alt = '<DEL>'
sv_type = 'DEL'
if type(sv_len) is int:
sv_len = -1 * sv_len
end = pos - sv_len
elif self.rearrangement == 'dup':
alt = '<DUP:TANDEM>'
sv_type = 'DUP'
elif self.rearrangement == 'inv':
alt = '<INV>'
sv_type = 'INV'
elif self.rearrangement == 'ins':
alt = '<INS>'
sv_type = 'INS'
end = pos
id = self.id if id_ext is None else id_ext
qual = '.'
filter = '.'
info = {
'SVTYPE': sv_type,
'END': end,
'BKPTID': ','.join(self.contigs),
}
if end is not None:
info['END'] = end
if type(sv_len) is int:
info['SVLEN'] = sv_len
if sv_type == 'DUP':
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not None:
info['FLANKING_PAIRS'] = self.support['flanking']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-':
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
#print 'gmap', contig_breaks
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
cipos = '0,%d' % homol_len
if cipos is not None:
info['CIPOS'] = cipos
info['CIPOS'] = cipos
if homol_len is not None:
info['HOMLEN'] = homol_len
info['HOMLEN'] = homol_len
if homol_seq is not None:
info['HOMSEQ'] = homol_seq
info['HOMSEQ'] = homol_seq
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if key == 'SVLEN' and value == 'NA':
continue
info[key] = value
if ref is not None and alt is not None:
fields = [
chrom, pos, id, ref, alt, qual, filter,
VCF.info_dict_to_str(info)
]
return '\t'.join(map(str, fields))
def as_breakends(self,
ref_fasta,
genomic=True,
max_novel_seq_len=50,
info_ext=None,
parids=None,
event=None):
chroms = map(lambda c: c.lstrip('chr'), self.chroms)
alt_chroms = chroms[:]
pos = list(self.breaks)
alt_pos = pos[:]
# inserted novel sequences
inserted_seqs = ['', '']
if self.novel_seq and self.novel_seq != 'NA' and self.novel_seq != '-':
if len(self.novel_seq) > max_novel_seq_len:
alt_chroms[0] = '<%s>' % self.contigs[0]
alt_chroms[1] = '<%s>' % self.contigs[0]
alt_pos[1] = self.contig_breaks[0][0] + 1
alt_pos[0] = self.contig_breaks[0][1] - 1
else:
if len(self.aligns[0]) == 1:
inserted_seqs[0] = self.novel_seq if self.aligns[0][
0].strand == '+' else reverse_complement(
self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][
0].strand == '+' else reverse_complement(
self.novel_seq)
else:
inserted_seqs[0] = self.novel_seq if self.aligns[0][
0].strand == '+' else reverse_complement(
self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][
1].strand == '+' else reverse_complement(
self.novel_seq)
# microhomology, cipos
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-' and len(
self.homol_seq) > 0:
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
pos[0] -= homol_len
alt_pos[1] += homol_len
cipos = '0,%d' % homol_len
refs = (ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[0]).upper(),
ref_fasta.fetch(self.chroms[1], self.breaks[1] - 1,
self.breaks[1]).upper())
ids = ('%s%s' % (self.id, 'a'), '%s%s' % (self.id, 'b'))
svtype = 'BND' if genomic else 'FND'
infos = [{
'SVTYPE': svtype,
'MATEID': ids[1],
'EVENTTYPE': self.rearrangement.upper()
}, {
'SVTYPE': svtype,
'MATEID': ids[0],
'EVENTTYPE': self.rearrangement.upper()
}]
if cipos is not None:
infos[0]['CIPOS'] = cipos
infos[1]['CIPOS'] = cipos
if homol_len is not None:
infos[0]['HOMLEN'] = homol_len
infos[1]['HOMLEN'] = homol_len
if homol_seq is not None:
infos[0]['HOMSEQ'] = homol_seq
infos[1]['HOMSEQ'] = homol_seq
# read support
if self.final_support is not None:
#infos[0]['READSUPPORT'] = self.final_support
#infos[1]['READSUPPORT'] = self.final_support
infos[0]['SPANNING_READS'] = self.support['spanning']
infos[1]['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not NONE:
infos[0]['FLANKING_PAIRS'] = self.support['flanking']
infos[1]['FLANKING_PAIRS'] = self.support['flanking']
adj_size = self.get_size()
if type(adj_size) is int:
infos[0]['SVLEN'] = adj_size
infos[1]['SVLEN'] = adj_size
# somatic
if self.somatic:
infos[0]['SOMATIC'] = 'SOMATIC'
infos[1]['SOMATIC'] = 'SOMATIC'
# contig and contig breakpoints
if self.contigs:
for i in range(2):
infos[i]['BKPTID'] = ','.join(self.contigs)
if self.contig_breaks and len(self.contig_breaks) == len(self.contigs):
contig_breaks = []
for bk in self.contig_breaks:
if len(bk) == 2:
contig_breaks.append('%s-%s' % (bk[0], bk[1]))
else:
print 'error'
if len(contig_breaks) == len(self.contigs):
for i in range(2):
infos[i]['CTG_BKS'] = ','.join(contig_breaks)
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if len(value) == 2:
infos[0][key] = value[0]
infos[1][key] = value[1]
if self.orients[0] == 'L':
# LL
if self.orients[1] == 'L':
alts = ('%s%s]%s:%s]' %
(refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
'%s%s]%s:%s]' %
(refs[1], inserted_seqs[1], alt_chroms[0], alt_pos[0]))
# LR
else:
alts = ('%s%s[%s:%s[' %
(refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
']%s:%s]%s%s' %
(alt_chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
else:
# RL
if self.orients[1] == 'L':
alts = (']%s:%s]%s%s' %
(chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'%s%s[%s:%s[' %
(refs[1], inserted_seqs[1], chroms[0], alt_pos[0]))
# RR
else:
alts = ('[%s:%s[%s%s' %
(chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'[%s:%s[%s%s' %
(chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
breakends = map(
lambda i: '\t'.join([
chroms[i],
str(pos[i]), ids[i], refs[i], alts[i], '.', '.',
VCF.info_dict_to_str(infos[i])
]), range(2))
return '\n'.join(breakends)
##indel_exonicfunc.xls
title_indel_exonicfunc = ['Sample','frameshift_deletion','frameshift_insertion','nonframeshift_deletion','nonframeshift_insertion','stoploss','stopgain','unknown']
indel_exonicfunc.write('\t'.join(title_indel_exonicfunc)+'\n')
for file in open(files, 'r'):
if file.startswith('#'):continue
file = file.strip()
sample_name = os.path.basename(file)
sample_name = sample_name.split('.')[0]
myVCF = VCF(file)
snp = myVCF.filter()
indel_file = file.replace('snp','indel')
myVCF = VCF(indel_file)
indel = myVCF.filter()
"""
##chr.xls
chr = myVCF.chr_stat(vcf)
chromosome.write(sample_name)
for i in [str(i) for i in range(1,23)]+['X','Y']:
try:
chromosome.write('\t'+str(chr[i]))
except:
chromosome.write('\t0')
chromosome.write('\n')
def as_breakends(self, ref_fasta, genomic=True, max_novel_seq_len=50, info_ext=None, parids=None, event=None):
chroms = map(lambda c: c.lstrip('chr'), self.chroms)
alt_chroms = chroms[:]
pos = list(self.breaks)
alt_pos = pos[:]
# inserted novel sequences
inserted_seqs = ['','']
if self.novel_seq and self.novel_seq != 'NA' and self.novel_seq != '-':
if len(self.novel_seq) > max_novel_seq_len:
alt_chroms[0] = '<%s>' % self.contigs[0]
alt_chroms[1] = '<%s>' % self.contigs[0]
alt_pos[1] = self.contig_breaks[0][0] + 1
alt_pos[0] = self.contig_breaks[0][1] - 1
else:
if len(self.aligns[0]) == 1:
inserted_seqs[0] = self.novel_seq if self.aligns[0][0].strand == '+' else reverse_complement(self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][0].strand == '+' else reverse_complement(self.novel_seq)
else:
inserted_seqs[0] = self.novel_seq if self.aligns[0][0].strand == '+' else reverse_complement(self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][1].strand == '+' else reverse_complement(self.novel_seq)
# microhomology, cipos
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-' and len(self.homol_seq) > 0:
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
pos[0] -= homol_len
alt_pos[1] += homol_len
cipos = '0,%d' % homol_len
refs = (ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[0]).upper(),
ref_fasta.fetch(self.chroms[1], self.breaks[1] - 1, self.breaks[1]).upper())
ids = ('%s%s' % (self.id, 'a'),
'%s%s' % (self.id, 'b'))
svtype = 'BND' if genomic else 'FND'
infos = [{'SVTYPE':svtype, 'MATEID':ids[1], 'EVENTTYPE':self.rearrangement.upper()},
{'SVTYPE':svtype, 'MATEID':ids[0], 'EVENTTYPE':self.rearrangement.upper()}]
if cipos is not None:
infos[0]['CIPOS'] = cipos
infos[1]['CIPOS'] = cipos
if homol_len is not None:
infos[0]['HOMLEN'] = homol_len
infos[1]['HOMLEN'] = homol_len
if homol_seq is not None:
infos[0]['HOMSEQ'] = homol_seq
infos[1]['HOMSEQ'] = homol_seq
# read support
if self.final_support is not None:
#infos[0]['READSUPPORT'] = self.final_support
#infos[1]['READSUPPORT'] = self.final_support
infos[0]['SPANNING_READS'] = self.support['spanning']
infos[1]['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not NONE:
infos[0]['FLANKING_PAIRS'] = self.support['flanking']
infos[1]['FLANKING_PAIRS'] = self.support['flanking']
adj_size = self.get_size()
if type(adj_size) is int:
infos[0]['SVLEN'] = adj_size
infos[1]['SVLEN'] = adj_size
# somatic
if self.somatic:
infos[0]['SOMATIC'] = 'SOMATIC'
infos[1]['SOMATIC'] = 'SOMATIC'
# contig and contig breakpoints
if self.contigs:
for i in range(2):
infos[i]['BKPTID'] = ','.join(self.contigs)
if self.contig_breaks and len(self.contig_breaks) == len(self.contigs):
contig_breaks = []
for bk in self.contig_breaks:
if len(bk) == 2:
contig_breaks.append('%s-%s' % (bk[0], bk[1]))
else:
print 'error'
if len(contig_breaks) == len(self.contigs):
for i in range(2):
infos[i]['CTG_BKS'] = ','.join(contig_breaks)
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if len(value) == 2:
infos[0][key] = value[0]
infos[1][key] = value[1]
if self.orients[0] == 'L':
# LL
if self.orients[1] == 'L':
alts = ('%s%s]%s:%s]' % (refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
'%s%s]%s:%s]' % (refs[1], inserted_seqs[1], alt_chroms[0], alt_pos[0]))
# LR
else:
alts = ('%s%s[%s:%s[' % (refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
']%s:%s]%s%s' % (alt_chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
else:
# RL
if self.orients[1] == 'L':
alts = (']%s:%s]%s%s' % (chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'%s%s[%s:%s[' % (refs[1], inserted_seqs[1], chroms[0], alt_pos[0]))
# RR
else:
alts = ('[%s:%s[%s%s' % (chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'[%s:%s[%s%s' % (chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
breakends = map(lambda i: '\t'.join([chroms[i], str(pos[i]), ids[i], refs[i], alts[i], '.', '.', VCF.info_dict_to_str(infos[i])]), range(2))
return '\n'.join(breakends)
def main():
input = ComLine(sys.argv[1:])
phy = Phylip(input.args.phy)
pops = Popmap(input.args.popmap)
VCF(phy, pops, input.args.out)
##indel_exonicfunc.xls
title_indel_exonicfunc = [
'Sample', 'frameshift_deletion', 'frameshift_insertion',
'nonframeshift_deletion', 'nonframeshift_insertion', 'stoploss',
'stopgain', 'unknown'
]
indel_exonicfunc.write('\t'.join(title_indel_exonicfunc) + '\n')
for file in open(files, 'r'):
if file.startswith('#'): continue
file = file.strip()
sample_name = os.path.basename(file)
sample_name = sample_name.split('.')[0]
myVCF = VCF(file)
snp = myVCF.filter()
indel_file = file.replace('snp', 'indel')
myVCF = VCF(indel_file)
indel = myVCF.filter()
"""
##chr.xls
chr = myVCF.chr_stat(vcf)
chromosome.write(sample_name)
for i in [str(i) for i in range(1,23)]+['X','Y']:
try:
chromosome.write('\t'+str(chr[i]))
except:
chromosome.write('\t0')
chromosome.write('\n')
"""