def __init__(self, output, header_str):
self.output = output
self.header_str = header_str
# create a cyvcf2 file for formatting, not for writing the file
tmp = tempfile.NamedTemporaryFile(mode="w", suffix=".vcf")
self.vcf = Writer.from_string(tmp, self.header_str)
# print the header
print(self.header_str, end="", file=self.output)
def to_vcf(self, path):
from cyvcf2 import Writer
header = '''##fileformat=VCFv4.2
#CHROM POS ID REF ALT QUAL FILTER INFO
'''
writer = Writer.from_string(path, header)
for v in self:
variant = writer.variant_from_string('\t'.join(
[v.chrom,
str(v.pos), '.', v.ref, v.alt, '.', '.', '.']))
writer.write_record(variant)
def test_writer_from_string():
header = """##fileformat=VCFv4.1
##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
##contig=<ID=chr2,length=249250621,assembly=hg19>
#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT samplea
"""
w = Writer.from_string("out.vcf", header)
w.write_header()
v = w.variant_from_string("chr1\t234\t.\tA\tC\t40\tPASS\t.\tGT\t0/0")
w.write_record(v)
w.close()
def to_vcf(self, path, remove_samples=False, clean_info=False):
"""
Parse query result as vcf file.
Args:
path: path of the file.
remove_samples: remove sample columns from vcf file
clean_info: clean info fields from vcf file
"""
from cyvcf2 import Writer
header = self.vcf.raw_header
# remove sample columns from header
# 2th last column in the header is columns
# all the columns after 8th is for samples
if remove_samples:
header = header.split('\n')
columns = header[-2].strip().split('\t')
header[-2] = '\t'.join(columns[:8])
header = '\n'.join(header)
writer = Writer.from_string(path, header)
for v in self:
variant = v.source
if remove_samples or clean_info:
variant = str(variant).strip().split()
if remove_samples:
# all the columns after 8th is for samples so remove them
variant = variant[:8]
if clean_info:
# 7th column contains info fields
# replace it with N/A
variant[7] = '.'
variant = writer.variant_from_string('\t'.join(variant))
writer.write_record(variant)
def main():
parser = argparse.ArgumentParser(description="""
Utility to convert bed-like file with alleles to a simple VCF
""")
parser.add_argument('-b',
'--bed',
help="""
bed file to convert to vcf
must have chrom, start, end, ref, and alt columns
""",
required=True)
parser.add_argument('-f',
'--fai',
help="""
fasta index file generated by samtools faidx
""",
required=True)
parser.add_argument('-o',
'--output',
help=""",
output file name,
prints to standard out if not supplied
""")
args = parser.parse_args()
# - is accepted as standard out for cyvcf2
if args.output is None:
output = "-"
else:
output = args.output
hdr = build_header(args.fai)
w = Writer.from_string(output, hdr)
bed_to_vcf(args.bed, w)
w.close()
# Getting the vcf header
rawheader = vcf.raw_header
newline = "#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\t%s"
headerlist = []
for line in rawheader.split('\n'):
if line.startswith('##'):
headerlist.append(line)
else:
pass
headerlist.append(newline)
header = '\n'.join(map(str, headerlist))
w_a = Writer.from_string(''.join([sample, '_a.vcf.gz']), header % sample_a)
w_a.write_header()
w_b = Writer.from_string(''.join([sample, '_b.vcf.gz']), header % sample_b)
w_b.write_header()
for v in vcf:
# class to get genotype in 0/0, 0/1, ... etc. format.
gts = v.genotypes
class Genotype(object):
__slots__ = ('alleles', 'phased')
def __init__(self, li):
self.alleles = li[:-1]
self.phased = li[-1]
