def write(self, sequence, selected_snv_location):
'It writes a seq with the alternative alleles in one position and Ns in the others.'
start = selected_snv_location - self._length
end = selected_snv_location + self._length + 1
if start < 0:
start = 0
if end > len(sequence):
end = len(sequence)
sequence = sequence[start: end]
selected_snv_location -= start
maf_threshold = self._maf
prev_seq_end = 0
seq_to_print = ''
for snv in sequence.get_features(kind='snv'):
# snv start and end [start, end[.
# Correcting the previous sequence slice
snv_start = snv.location.start.position - start
snv_end = snv.location.end.position - start
# join the previous sequence to the sequence to print
seq_to_print += str(sequence[prev_seq_end:snv_start].seq)
prev_seq_end = snv_end
if snv_start == selected_snv_location:
#subtituir por allelos
snv_kind = calculate_snv_kind(snv)
if snv_kind != SNP:
msg = "We don't know how to print anything but SNPs"
raise NotImplementedError(msg)
alleles = '/'.join([a[0] for a in snv.qualifiers['alleles'].keys()])
to_print = '[{0:s}]'.format(alleles)
else:
if maf_threshold is not None:
snv_maf = calculate_maf_frequency(snv)
write_abundant_allele = True if snv_maf > maf_threshold else False
else:
write_abundant_allele = False
if write_abundant_allele:
# most abundant allele
to_print = _get_major_allele(snv)
else:
# Ns
snv_kind = calculate_snv_kind(snv)
if snv_kind == SNP:
to_print = _snp_to_iupac(snv, sequence)
elif snv_kind in (DELETION, COMPLEX, INDEL):
ref_allele = snv.qualifiers['reference_allele']
to_print = ref_allele[0] + 'N' * (len(ref_allele) - 1)
else:
to_print = 'N'
seq_to_print += to_print
else:
seq_to_print += str(sequence[prev_seq_end:end + 1].seq)
name = sequence.name + '_' + str(selected_snv_location + 1)
self.fhand.write('>%s\n%s\n' % (name, seq_to_print))
self.fhand.flush()
def _snv_to_string(snv, sequence, position):
'''it writes the snv in a format with braquets and with all alleles:
[A/T], [-/ATG]...'''
snv_kind = calculate_snv_kind(snv)
reference_allele = snv.qualifiers['reference_allele']
alleles = [allele[0] for allele in snv.qualifiers['alleles'].keys() if allele[1] != INVARIANT]
alleles = set(alleles)
assert reference_allele.upper() in ['A', 'T', 'C', 'G']
if snv_kind == SNP:
to_print = "%s/%s" % (reference_allele, "/".join(alleles))
size = 1
elif snv_kind == COMPLEX:
raise RuntimeError('Complex type not implemented')
elif snv_kind == INDEL:
if len(alleles) > 1:
raise RuntimeError('INDEL type not implemented')
elif snv_kind == INSERTION:
allele = list(alleles)[0]
size = 1
to_print = '-/' + allele
elif snv_kind == DELETION:
allele = list(alleles)[0]
size = len(allele)
deleted_alleles = sequence[position:position+size]
to_print = '%s/-' % ''.join(deleted_alleles)
to_print = '[%s]' % to_print
return to_print, size
def kind_filter(sequence):
'The filter'
if sequence is None:
return None
for snv in sequence.get_features(kind='snv'):
previous_result = _get_filter_result(snv, 'by_kind', threshold=kind)
if previous_result is not None:
continue
kind_ = calculate_snv_kind(snv)
if kind != kind_:
result = True
else:
result = False
_add_filter_result(snv, 'by_kind', result, threshold=kind)
return sequence
def _do_snv_stats(stats, feats, orfs):
"It adds the ssr stats"
some_feat = False
for feat in feats:
some_feat = True
type_ = calculate_snv_kind(feat, detailed=True)
if type_ not in stats["snvs"]["types"]:
stats["snvs"]["types"][type_] = 0
stats["snvs"]["types"][type_] += 1
location = _location_to_orf(orfs, feat)
if location:
if location not in stats["snvs"]["locations"]:
stats["snvs"]["locations"][location] = 0
stats["snvs"]["locations"][location] += 1
stats["snvs"]["n_snvs"] += 1
if some_feat:
stats["snvs"]["n_seqs"] += 1
def _do_snv_stats(stats, feats, orfs):
'It adds the ssr stats'
some_feat = False
for feat in feats:
some_feat = True
type_ = calculate_snv_kind(feat, detailed=True)
if type_ not in stats['snvs']['types']:
stats['snvs']['types'][type_] = 0
stats['snvs']['types'][type_] += 1
location = _location_to_orf(orfs, feat)
if location:
if location not in stats['snvs']['locations']:
stats['snvs']['locations'][location] = 0
stats['snvs']['locations'][location] += 1
stats['snvs']['n_snvs'] += 1
if some_feat:
stats['snvs']['n_seqs'] += 1
