def extract_pharmcat_pgx_regions(tabix_executable_path, input_vcf, output_dir,
input_ref_pgx_vcf):
'''
extract pgx regions in input_ref_pgx_vcf from input_vcf and save variants to path_output
'''
print(
'Modify chromosome names.\nExtract PGx regions based on the input reference PGx position file.'
)
path_output = os.path.join(
output_dir,
obtain_vcf_file_prefix(input_vcf) + '.pgx_regions.vcf.gz')
input_vcf_cyvcf2 = VCF(input_vcf)
input_ref_pgx_pos_cyvcf2 = VCF(input_ref_pgx_vcf)
# get pgx regions in each chromosome
input_ref_pgx_pos_pandas = allel.vcf_to_dataframe(input_ref_pgx_vcf)
input_ref_pgx_pos_pandas['CHROM'] = input_ref_pgx_pos_pandas[
'CHROM'].replace({
'chr': ''
}, regex=True).astype(str).astype(int)
ref_pgx_regions = input_ref_pgx_pos_pandas.groupby(
['CHROM'])['POS'].agg(get_vcf_pos_min_max).reset_index()
# fix chr names
chr_name_match = re.compile("^chr")
if any(chr_name_match.match(line) for line in input_vcf_cyvcf2.seqnames):
# chromosomes have leading 'chr' characters in the original VCF
# pgx regions to be extracted
ref_pgx_regions = ref_pgx_regions.apply(
lambda row: ':'.join(row.values.astype(str)),
axis=1).replace({'^': 'chr'}, regex=True)
else:
# chromosomes do not have leading 'chr' characters in the original VCF
# add chromosome name with leading 'chr' to the VCF header
for single_chr in input_vcf_cyvcf2.seqnames:
input_vcf_cyvcf2.add_to_header('##contig=<ID=chr' + single_chr +
'>')
# pgx regions to be extracted
ref_pgx_regions = ref_pgx_regions.apply(
lambda row: ':'.join(row.values.astype(str)), axis=1)
# write to a VCF output file
# header
output_vcf_cyvcf2 = Writer(path_output, input_vcf_cyvcf2, mode="wz")
# content
for single_region in ref_pgx_regions:
for single_variant in input_vcf_cyvcf2(single_region):
single_variant.CHROM = re.sub(r'^([0-9]+)', r'chr\1',
single_variant.CHROM)
output_vcf_cyvcf2.write_record(single_variant)
# close pipe
input_vcf_cyvcf2.close()
input_ref_pgx_pos_cyvcf2.close()
output_vcf_cyvcf2.close()
tabix_index_vcf(tabix_executable_path, path_output)
return path_output
def annotate(vcf, roh, bed, v14, quality_threshold, flag_upd_at_fraction,
output, verbose):
"""Markup VCF file using rho-calls. Use BED file to mark all variants in AZ
windows. Alternatively, use a bcftools v>=1.4 file with RG entries to mark
all vars. With the --no-v14 flag, use an older bcftools v<=1.2 style roh TSV
to mark only selected AZ variants. Roh is broken in bcftools v1.3
- do not use."""
loglevel = LEVELS.get(min(verbose, 3))
configure_stream(level=loglevel)
proband_vcf = VCF(vcf)
# add this command to VCF header
## This is for logging the command line string ##
frame = inspect.currentframe()
args, _, _, values = inspect.getargvalues(frame)
argument_list = [
i + '=' + str(values[i]) for i in values
if values[i] and i not in ['frame']
]
logger.info("Running rhocall annotate {0}".format(__version__))
logger.debug("Arguments: {0}".format(', '.join(argument_list)))
## add additional tags to VCF header
proband_vcf.add_to_header('##rhocall_version={0}'.format(__version__))
proband_vcf.add_to_header("##rhocall_arguments={0}".format(
', '.join(argument_list)))
if roh and not bed and not v14:
run_annotate_var(proband_vcf=proband_vcf,
roh=roh,
quality_threshold=quality_threshold,
flag_upd_at_fraction=flag_upd_at_fraction,
output=output)
elif roh and v14 and not bed:
run_annotate_rg(proband_vcf=proband_vcf,
bcfroh=roh,
quality_threshold=quality_threshold,
flag_upd_at_fraction=flag_upd_at_fraction,
output=output)
elif bed and not roh:
run_annotate(proband_vcf=proband_vcf,
bed=bed,
quality_threshold=quality_threshold,
flag_upd_at_fraction=flag_upd_at_fraction,
output=output)
else:
click.echo("""Cannot use both BED and ROH at once. Please apply
them sequentially instead.""")
raise ValueError(f"{fai} does not exist.")
TEMPLATE_VCF = f"""##fileformat=VCFv4.2
##FILTER=<ID=PASS,Description="All filters passed">
##fileDate={date.today().strftime("%Y%m%d")}
#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\t{sample}
chr1\t1\t.\tA\tT\t.\tPASS\t.\tGT\t0|0
"""
TEMPLATE_VCF_FILE = joboutdir / "template.vcf"
TEMPLATE_VCF_FILE.write_text(TEMPLATE_VCF)
vcf = VCF(TEMPLATE_VCF_FILE)
# Add source
vcf.add_to_header(f"##source=biopipen.ns.bed.Bed2Vcf")
# Add genome assembly
if genome:
vcf.add_to_header(f"##reference={genome}")
vcf.add_info_to_header(
{
"ID": "END",
"Number": "1",
"Type": "Integer",
"Description": "End position of the variant described in this record"
}
)
vcf.add_format_to_header(
logger.success(f"Extracted {len(genes)} genes from the panel")
logger.info("Extracting intervals for genes from GFF...")
with open(snakemake.input.annotation) as istream:
ivtree = extract_intervals_for_genes_from_gff(genes, istream, padding)
logger.success(f"Intervals extracted for {len(ivtree)} genes")
logger.info(
"Extracting those VCF records that fall within the gene intervals and altering "
"their CHROM and POS accordingly...")
vcf_reader = VCF(snakemake.input.vcf)
logger.debug("Adding genes to header...")
for iv in ivtree:
vcf_reader.add_to_header(
f"##contig=<ID={iv.data[0]},length={iv.end-iv.begin}>")
logger.debug("Genes added to header")
with TemporaryDirectory() as tmpdirname:
tmpvcf = str(Path(tmpdirname) / "tmp.vcf")
vcf_writer = Writer(tmpvcf, tmpl=vcf_reader)
for record in vcf_reader:
if apply_filters and record.FILTER is not None:
continue
gt = Genotype.from_arr(record.genotypes[0])
if only_alt and not gt.is_hom_alt():
continue
ivs = ivtree[record.start]
