def tab_to_vcf(input_file, output_file, reference_file, columns, info_fields, convert_iupac=False):
"""
Convert tab-delimited file to VCF.
Support for the fixed VCF fields: #CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO
PyVCF's _Record class requires the following arguments:
CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO, FORMAT, sample_indexes
convert_iupac (bool) : When present, convert IUPAC codes to the non-reference allele.
This is only possible for when the reference and IUPAC-determined alternates share
at least one allele. Tri-allelic conversion is not supported and will emit a warning.
IUPAC codes: http://www.bioinformatics.org/sms/iupac.html
"""
reference_dict = FastaHack(reference_file)
with open(input_file, "r") as input_fh:
reader = csv.DictReader(input_fh, delimiter="\t")
with open(TEMPLATE_VCF_FILE, "r") as template_fh:
vcf_reader = vcf.Reader(template_fh)
with open(output_file, "w") as output_fh:
vcf_writer = vcf.Writer(output_fh, vcf_reader, lineterminator='\n')
for row in reader:
args = [row.get(columns.get(f,None), ".") for f in VCF_COLUMN_ORDER]
# Convert position to an integer.
args[POSITION_INDEX] = int(args[POSITION_INDEX])
# Convert indels from GATK to VCF format.
if args[ALT_INDEX].startswith(("+", "-")) and not "/" in args[ALT_INDEX]:
args = gatk_indel_to_vcf(args, reference_dict)
# Optionally convert IUPAC code
if convert_iupac:
args = _convert_iupac(args)
# Convert alternate allele scalar to a list.
args[ALT_INDEX] = [args[ALT_INDEX]]
# Convert info fields
if info_fields:
INFO = {}
for vcf_field,tab_field in info_fields.items():
if tab_field in row:
INFO[vcf_field] = row[tab_field]
else:
INFO = {}
# Add empty entries for INFO, FORMAT, and sample_indexes.
args.extend([INFO, ".", []])
record = _Record(*args)
vcf_writer.write_record(record)
def tab_to_vcf(input_file, output_file, reference_file):
"""
Convert tab-delimited file to VCF.
Support for the fixed VCF fields: #CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO
PyVCF's _Record class requires the following arguments:
CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO, FORMAT, sample_indexes
"""
reference_dict = FastaHack(reference_file)
with open(input_file, "r") as input_fh:
reader = csv.DictReader(input_fh, delimiter="\t")
with open(TEMPLATE_VCF_FILE, "r") as template_fh:
vcf_reader = vcf.Reader(template_fh)
with open(output_file, "w") as output_fh:
vcf_writer = vcf.Writer(output_fh,
vcf_reader,
lineterminator='\n')
for row in reader:
args = [
row.get(tab_field, ".")
for vcf_field, tab_field in VCF_TO_FIELDS
]
# Convert position to an integer.
args[POSITION_INDEX] = int(args[POSITION_INDEX])
# Convert indels from GATK to VCF format.
if args[ALT_INDEX].startswith(
("+", "-")) and not "/" in args[ALT_INDEX]:
args = gatk_indel_to_vcf(args, reference_dict)
# Convert alternate allele scalar to a list.
args[ALT_INDEX] = [args[ALT_INDEX]]
# Add empty entries for INFO, FORMAT, and sample_indexes.
args.extend([{}, ".", []])
record = _Record(*args)
vcf_writer.write_record(record)
