def add_chunk(self, sample, pred):
# Write consensus alts to vcf
cursor = 0
var_queue = list()
ref_seq = self.ref_fasta.fetch(sample.ref_name)
for pos, grp in itertools.groupby(sample.positions['major']):
end = cursor + len(list(grp))
alt = ''.join(self.label_decoding[x]
for x in pred[cursor:end]).replace(_gap_, '')
# For simple insertions and deletions in which either
# the REF or one of the ALT alleles would otherwise be
# null/empty, the REF and ALT Strings must include the
# base before the event (which must be reflected in
# the POS field), unless the event occurs at position
# 1 on the contig in which case it must include the
# base after the event
if alt == '':
# deletion
if pos == 0:
# the "unless case"
ref = ref_seq[1]
alt = ref_seq[1]
else:
# the usual case
pos = pos - 1
ref = ref_seq[pos:pos + 2]
alt = ref_seq[pos]
else:
ref = ref_seq[pos]
# Merging of variants produced by considering major.{minor} positions
# These are of the form:
# X -> Y - subs
# prev.X -> prev - deletion
# X -> Xyy.. - insertion
# In the second case we may need to merge variants from consecutive
# major positions.
if alt == ref:
self.write(var_queue)
var_queue = list()
else:
var = vcf.Variant(sample.ref_name, pos, ref, alt)
if len(var_queue) == 0 or pos - var_queue[-1].pos == 1:
var_queue.append(var)
else:
self.write(var_queue)
var_queue = [var]
cursor = end
self.write(var_queue)
def write(self, var_queue):
if len(var_queue) > 1:
are_dels = all(len(x.ref) == 2 for x in var_queue)
are_same_ref = len(set(x.chrom for x in var_queue)) == 1
if are_dels and are_same_ref:
name = var_queue[0].chrom
pos = var_queue[0].pos
ref = ''.join((x.ref[0] for x in var_queue))
ref += var_queue[-1].ref[-1]
alt = ref[0]
merged_var = vcf.Variant(name, pos, ref, alt)
self.writer.write_variant(merged_var)
else:
raise ValueError('Cannot merge variants: {}.'.format(var_queue))
elif len(var_queue) == 1:
self.writer.write_variant(var_queue[0])
def write(self, vcf_fh):
if len(self) > 1:
are_dels = all(len(x.ref) == 2 for x in self)
are_same_ref = len(set(x.chrom for x in self)) == 1
if are_dels and are_same_ref:
name = self[0].chrom
pos = self[0].pos
ref = ''.join((x.ref[0] for x in self))
ref += self[-1].ref[-1]
alt = ref[0]
merged_var = vcf.Variant(name, pos, ref, alt, info=info)
vcf_fh.write_variant(merged_var)
else:
raise ValueError('Cannot merge variants: {}.'.format(self))
elif len(self) == 1:
vcf_fh.write_variant(self[0])
del self[:]