def parse_vcf(vcffile, logfh):
"""Parse sample VCF file."""
logfh.write("Parsing vcffile: {} -".format(vcffile))
fh = open_file(vcffile)
lines = fh.readlines()
fh.close()
while lines: # skip vcf header
if not lines[0].startswith('##'):
break
lines.pop(0)
fields = lines.pop(0).lstrip('#').rstrip().split("\t")
variants = {}
for line in lines:
vals = line.rstrip().split("\t")
d = dict(zip(fields, vals))
if d['QUAL'][0].isdigit():
pos = int(d['POS'])
variants[pos] = d
logfh.write(" {} variants\n".format(len(variants)))
# special case for counting (TG)9-9T
# if 87842 in variants and 87823 not in variants:
# variants[87823] = {'CHROM':'CFTR', 'POS':'87823',
# 'REF':'ATG', 'ALT':'A' }
# if 87844 in variants and 87845 not in variants:
# variants[87845] = {'CHROM':'CFTR', 'POS':'87845',
# 'REF':'G', 'ALT':'T' }
# special case for counting (TG)13-5T
# if 87846 in variants and 87823 not in variants:
# variants[87823] = {'CHROM':'CFTR', 'POS':'87823',
# 'REF':'A', 'ALT':'ATG' }
# if 87847 in variants and variants[87847]['REF']=='T' and\
# variants[87847]['ALT']=='TG':
# variants[87847]['ALT'] += ',G'
return (fields, variants)
def separate_vcf(vcffile, ref, outdir, args):
sys.stderr.write("\nReading {}\n".format(vcffile))
fh = open_file(vcffile)
fieldlist = [ l for l in fh.readlines() if l.startswith('#CHROM') ]
fh.close()
if fieldlist:
fields = fieldlist[0].rstrip().split("\t")
samples = fields[9:]
else:
sys.stderr.write("Could not find VCF header in {}".format(vcffile))
sys.exit(1)
sys.stderr.write(" Found {} samples.\n".format(len(samples)))
outvcfs = []
logfh = sys.stderr
if args.logdir:
logfile = os.path.basename(vcffile).replace('.vcf','') +\
".separate_vcf.log"
logfile = os.path.join(args.logdir, logfile)
logfh = open(logfile, 'w')
for sample in samples:
outvcf = "{}.separated.vcf".format(sample)
if outdir: outvcf = os.path.join(outdir, outvcf)
run_select_variants(vcffile, outvcf, sample, ref, logfh, args)
if args.logdir:
logfh.close()
return samples
def parse_vcf(vcffile):
"""Parse variants in joint vcf file."""
sys.stderr.write("VCF file {} -".format(vcffile))
fh = open_file(vcffile)
lines = fh.readlines()
fh.close()
while lines: # skip vcf header
if not lines[0].startswith('##'):
break
lines.pop(0)
fields = lines.pop(0).lstrip('#').rstrip().split("\t")
variants = []
for line in lines:
vals = line.rstrip().split("\t")
d = dict(zip(fields, vals))
variants.append(d)
sys.stderr.write(" {} variants\n".format(len(variants)))
return (fields, variants)
def parse_vcf(vcffile):
"""Parse variants in joint vcf file."""
sys.stderr.write("VCF file {} -".format(vcffile))
fh = open_file(vcffile)
lines = fh.readlines()
fh.close()
while lines: # skip vcf header
if not lines[0].startswith('##'):
break
lines.pop(0)
fields = lines.pop(0).lstrip('#').rstrip().split("\t")
variants = []
for line in lines:
vals = line.rstrip().split("\t")
d = dict(zip(fields, vals))
variants.append(d)
sys.stderr.write(" {} variants\n".format(len(variants)))
return (fields, variants)
def parse_vcf(vcffile):
fh = open_file(vcffile)
lines = fh.readlines()
fh.close()
header = []
while lines: # vcf header
if lines[0].startswith('#'):
header.append(lines.pop(0))
else:
break
fields = header[-1].lstrip('#').rstrip().split("\t")
vcfinfo = {}
for line in lines:
vals = line.rstrip().split("\t")
d = dict(zip(fields, vals))
poskey = create_pos_key(d['CHROM'], d['POS'])
if poskey in vcfinfo:
sys.stderr.write(" Variant pos {} duplicated\n".format(poskey))
vcfinfo[poskey] = d
return (header, fields, vcfinfo)
def parse_vcf(vcffile):
fh = open_file(vcffile)
lines = fh.readlines()
fh.close()
header = []
while lines: # vcf header
if lines[0].startswith('#'):
header.append(lines.pop(0))
else:
break
fields = header[-1].lstrip('#').rstrip().split("\t")
vcfinfo = {}
for line in lines:
vals = line.rstrip().split("\t")
d = dict(zip(fields, vals))
poskey = create_pos_key(d['CHROM'], d['POS'])
if poskey in vcfinfo:
sys.stderr.write(" Variant pos {} duplicated\n".format(poskey))
vcfinfo[poskey] = d
return (header, fields, vcfinfo)
def parse_bedfile(bedfile):
"""Parse ROI from bed file."""
sys.stderr.write("\nReading {}\n".format(bedfile))
fh = open_file(bedfile)
lines = fh.readlines()
fh.close()
roi = defaultdict(dict)
fields = ['CHROM', 'START', 'END', 'NAME']
numlines = 0
for line in lines:
vals = line.rstrip().split("\t")
d = dict(zip(fields, vals))
d['START'] = int(d['START']) + 1 # convert to 1-based
d['END'] = int(d['END'])
roi[d['CHROM']][d['START']] = d
numlines += 1
sys.stderr.write(" Got {} lines in {} chrom\n".format(numlines,
len(roi.keys())))
return roi
def trim_primers(fqfile, alignout, max_trim_len, primerinfo, outlabel, logfh,
args):
"""Returns trimmed fastq file and file with list of sequence names"""
trimmedfq = outlabel + ".trimmed.fastq"
seqfile = outlabel + ".seqlist.txt"
logfh.write(" Trimming fq: {}\n".format(trimmedfq))
if have_files([trimmedfq, seqfile], args.force, stderr=logfh):
logfh.write(" Already have {}\n".format(trimmedfq))
return (trimmedfq, seqfile)
aligns = parse_alignout(alignout)
seqlist = []
with open(trimmedfq, 'w') as outfq:
inseq = FastQParser(fqfile)
for seqrec in inseq:
seqlist.append(seqrec.id)
if seqrec.id in aligns:
primer = aligns[seqrec.id]['primer']
if primerinfo[primer]['overlap']:
primerend = aligns[seqrec.id]['end'] +\
aligns[seqrec.id]['left']
subrec = seqrec[primerend:]
if args.debug:
logfh.write("{}\tTrimming\t{}\n".format(
primer, seqrec.id))
else:
if args.debug:
logfh.write("{}\tNot trimming\t{}\n".format(
primer, seqrec.id))
subrec = seqrec
else: #trim default max_primer_len+2
subrec = seqrec[max_trim_len:]
outfq.write("{}\n".format(subrec.fastq()))
logfh.write(" Seq list: {}\n".format(seqfile))
if have_file(seqfile, True, stderr=logfh):
logfh.write(" Still have {}\n".format(seqfile))
sys.exit()
with open_file(seqfile, 'w') as ifh:
ifh.write("\n".join(seqlist) + "\n")
return (trimmedfq, seqfile)
def trim_primers(fqfile, alignout, max_trim_len, primerinfo, outlabel, logfh,
args):
"""Returns trimmed fastq file and file with list of sequence names"""
trimmedfq = outlabel + ".trimmed.fastq"
seqfile = outlabel + ".seqlist.txt"
logfh.write(" Trimming fq: {}\n".format(trimmedfq))
if have_files([trimmedfq, seqfile], args.force, stderr=logfh):
logfh.write(" Already have {}\n".format(trimmedfq))
return (trimmedfq, seqfile)
aligns = parse_alignout(alignout)
seqlist = []
with open(trimmedfq, 'w') as outfq:
inseq = FastQParser(fqfile)
for seqrec in inseq:
seqlist.append(seqrec.id)
if seqrec.id in aligns:
primer = aligns[seqrec.id]['primer']
if primerinfo[primer]['overlap']:
primerend = aligns[seqrec.id]['end'] +\
aligns[seqrec.id]['left']
subrec = seqrec[primerend:]
if args.debug:
logfh.write("{}\tTrimming\t{}\n".format(
primer, seqrec.id))
else:
if args.debug:
logfh.write("{}\tNot trimming\t{}\n".format(
primer, seqrec.id))
subrec = seqrec
else: #trim default max_primer_len+2
subrec = seqrec[max_trim_len:]
outfq.write("{}\n".format(subrec.fastq()))
logfh.write(" Seq list: {}\n".format(seqfile))
if have_file(seqfile, True, stderr=logfh):
logfh.write(" Still have {}\n".format(seqfile))
sys.exit()
with open_file(seqfile, 'w') as ifh:
ifh.write("\n".join(seqlist)+"\n")
return (trimmedfq, seqfile)