def get_variant_from_vcf_fields(vcf_fields, alt_allele_pos):
"""
Get a basic variant from vcf_fields, for allele given by alt_allele_pos
"""
chrom = vcf_fields[0] if 'chr' in vcf_fields[0] else 'chr' + vcf_fields[0]
pos = int(vcf_fields[1])
# if we can't get a genomic location, just ignore it and print a message humans will ignore too
# obviously need a better way to approach this
if not genomeloc.valid_pos(chrom, pos):
print "ERROR: could not figure out coordinates for %s:%d...maybe a nonstandard chromosome?" % (chrom, pos)
return None
ref = vcf_fields[3]
orig_alt_alleles = vcf_fields[4].split(',')
alt = orig_alt_alleles[alt_allele_pos]
xpos = genomeloc.get_single_location(chrom, pos)
xpos, ref, alt = get_minimal_representation(xpos, ref, alt)
variant = Variant(xpos, ref, alt)
variant.set_extra('alt_allele_pos', alt_allele_pos)
variant.set_extra('orig_alt_alleles', orig_alt_alleles)
if vcf_fields[2] and vcf_fields[2] != '.':
variant.vcf_id = vcf_fields[2]
return variant
def get_variant_from_vcf_fields(vcf_fields, alt_allele_pos):
"""
Get a basic variant from vcf_fields, for allele given by alt_allele_pos
"""
chrom = vcf_fields[0] if 'chr' in vcf_fields[0] else 'chr' + vcf_fields[0]
pos = int(vcf_fields[1])
# if we can't get a genomic location, just ignore it and print a message humans will ignore too
# obviously need a better way to approach this
if not genomeloc.valid_pos(chrom, pos):
print "ERROR: could not figure out coordinates for %s:%d...maybe a nonstandard chromosome?" % (
chrom, pos)
return None
ref = vcf_fields[3]
orig_alt_alleles = vcf_fields[4].split(',')
alt = orig_alt_alleles[alt_allele_pos]
xpos = genomeloc.get_single_location(chrom, pos)
xpos, ref, alt = get_minimal_representation(xpos, ref, alt)
variant = Variant(xpos, ref, alt)
variant.set_extra('alt_allele_pos', alt_allele_pos)
variant.set_extra('orig_alt_alleles', orig_alt_alleles)
if vcf_fields[2] and vcf_fields[2] != '.':
variant.vcf_id = vcf_fields[2]
return variant
def get_exac_af(chrom, pos, ref, alt):
populations = ['AMR', 'EAS', 'FIN', 'NFE', 'SAS', 'AFR']
chrom_without_chr = chrom.replace("chr", "")
xpos = genomeloc.get_single_location(chrom, pos)
variant_length = len(ref) + len(alt)
# check whether the alleles match
matching_exac_variant = None
matching_exac_variant_i = None
for record in exac_vcf.fetch(chrom_without_chr, pos - variant_length,
pos + variant_length):
exac_xpos = genomeloc.get_xpos(record.CHROM, record.POS)
for exac_alt_i, exac_alt in enumerate(record.ALT):
exac_variant_xpos, exac_ref, exac_alt = get_minimal_representation(
exac_xpos, str(record.REF), str(exac_alt))
if exac_variant_xpos == xpos and exac_ref == ref and exac_alt == alt:
if matching_exac_variant is not None:
print(
"ERROR: multiple exac variants match the variant: %s %s %s %s"
% (chrom, pos, ref, alt))
matching_exac_variant = record
matching_exac_variant_i = exac_alt_i
#print("Variant %s %s %s matches: %s %s %s %s" % (xpos, ref, alt, record, exac_variant_xpos, exac_ref, exac_alt) )
if matching_exac_variant is None:
#print("Variant %s %s %s %s not found in ExAC" % (chrom, pos, alt, ref))
return None, None, None
pop_max_af = -1
pop_max_population = None
for p in populations:
if matching_exac_variant.INFO['AN_' + p] > 0:
pop_af = matching_exac_variant.INFO[
'AC_' + p][matching_exac_variant_i] / float(
matching_exac_variant.INFO['AN_' + p])
if pop_af > pop_max_af:
pop_max_af = pop_af
pop_max_population = p
if matching_exac_variant.INFO['AN_Adj'] != 0:
global_af = float(matching_exac_variant.INFO['AC_Adj']
[matching_exac_variant_i]) / float(
matching_exac_variant.INFO['AN_Adj'])
else:
assert float(
matching_exac_variant.INFO['AC_Adj'][matching_exac_variant_i]) == 0
global_af = 0
return global_af, pop_max_af, pop_max_population
def get_exac_af(chrom, pos, ref, alt):
populations = ['AMR', 'EAS', 'FIN', 'NFE', 'SAS', 'AFR']
chrom_without_chr = chrom.replace("chr", "")
xpos = genomeloc.get_single_location(chrom, pos)
variant_length = len(ref) + len(alt)
# check whether the alleles match
matching_exac_variant = None
matching_exac_variant_i = None
for record in exac_vcf.fetch(chrom_without_chr, pos - variant_length, pos + variant_length):
exac_xpos = genomeloc.get_xpos(record.CHROM, record.POS)
for exac_alt_i, exac_alt in enumerate(record.ALT):
exac_variant_xpos, exac_ref, exac_alt = get_minimal_representation(exac_xpos, str(record.REF), str(exac_alt))
if exac_variant_xpos == xpos and exac_ref == ref and exac_alt == alt:
if matching_exac_variant is not None:
print("ERROR: multiple exac variants match the variant: %s %s %s %s" % (chrom, pos, ref, alt))
matching_exac_variant = record
matching_exac_variant_i = exac_alt_i
#print("Variant %s %s %s matches: %s %s %s %s" % (xpos, ref, alt, record, exac_variant_xpos, exac_ref, exac_alt) )
if matching_exac_variant is None:
#print("Variant %s %s %s %s not found in ExAC" % (chrom, pos, alt, ref))
return None, None, None
pop_max_af = -1
pop_max_population = None
for p in populations:
if matching_exac_variant.INFO['AN_'+p] > 0:
pop_af = matching_exac_variant.INFO['AC_'+p][matching_exac_variant_i]/float(matching_exac_variant.INFO['AN_'+p])
if pop_af > pop_max_af:
pop_max_af = pop_af
pop_max_population = p
if matching_exac_variant.INFO['AN_Adj'] != 0:
global_af = float(matching_exac_variant.INFO['AC_Adj'][matching_exac_variant_i])/float(matching_exac_variant.INFO['AN_Adj'])
else:
assert float(matching_exac_variant.INFO['AC_Adj'][matching_exac_variant_i]) == 0
global_af = 0
return global_af, pop_max_af, pop_max_population
