def find_variations_md(cigartuples, chromosome, pos, qual, seq, md, subsegments):
"""
Loops through read to find variant base. This function is used with BWA-MEM, Minimap2 and NGMLR bam files
:param cigartuples:
:param chromosome:
:param pos:
:param qual:
:param seq:
:param md:
:param subsegments:
"""
global tmp_variants
md_tag = re.split("\^\D+", md)
read_cursor = 0
ref_cursor = 0
md_cursor = 0
md_string = ''
for tuple in cigartuples:
if tuple[0] == 4:
read_cursor += tuple[1]
if tuple[0] == 1:
read_cursor += tuple[1]
if tuple[0] == 2:
md_cursor += 1
md_string = ''
for deletion in range(tuple[1]):
if (pos + 1 + ref_cursor) not in tmp_variants:
tmp_variants[pos + 1 + ref_cursor] = v.Variant(chromosome, pos + 1 + ref_cursor)
for segment in subsegments:
if segment.pos <= (pos + 1 + ref_cursor) <= segment.end:
tmp_variants[pos + 1 + ref_cursor].add_segment(segment.id, ['-'])
break
if segment.end < ref_cursor:
del subsegments[subsegments.index(segment)]
ref_cursor += 1
if tuple[0] == 0:
if re.search("\D+", md_tag[md_cursor]):
if md_string == '':
for m in re.split("\D", md_tag[md_cursor]):
if md_string != '':
md_string += 'X'
md_string += "=" * int(m)
for seq_mismatch in re.finditer('X', md_string[:tuple[1]]):
if (pos + 1 + ref_cursor) not in tmp_variants:
tmp_variants[pos + 1 + ref_cursor] = v.Variant(chromosome, pos + 1 + ref_cursor)
for segment in subsegments:
if segment.pos <= (pos + 1 + ref_cursor) <= segment.end:
tmp_variants[pos + 1 + ref_cursor].add_segment(segment.id, [seq[read_cursor + seq_mismatch.start()], qual[read_cursor + seq_mismatch.start()]])
break
if segment.end < ref_cursor:
del subsegments[subsegments.index(segment)]
md_string = md_string[tuple[1]:]
ref_cursor += tuple[1]
read_cursor += tuple[1]
def find_SNPs(chromosome, snp_position):
"""
Looks for SNPs on given position in the genome using pileup.
:param chromosome:
:param snp_position:
"""
base_ratios = {'A': [0, 0], 'C': [0, 0], 'G': [0, 0], 'T': [0, 0], '=': [0, 0]}
deletions = 0
total_n = 0
variant = v.Variant(chromosome, int(snp_position))
for pileupcolumn in F.pileup(chromosome, int(snp_position)-1, int(snp_position), truncate=True):
for pileupread in pileupcolumn.pileups:
if not keep_segment(pileupread.alignment, pileupread.alignment.query_alignment_length):
continue
clip, clip_2 = calculate_clip(pileupread.alignment)
if pileupread.is_del:
variant.add_segment([pileupread.alignment.reference_name, pileupread.alignment.reference_start, str(pileupread.alignment.query_name) + ";" + str(clip)], '-')
deletions += 1
total_n += 1
if not pileupread.is_del and not pileupread.is_refskip:
if pileupread.alignment.query_qualities[pileupread.query_position] >= NanoSV.opts_min_base_qual_ph:
base_ratios[pileupread.alignment.query_sequence[pileupread.query_position]][0] += 1
else:
base_ratios[pileupread.alignment.query_sequence[pileupread.query_position]][1] += 1
variant.add_segment([pileupread.alignment.reference_name, pileupread.alignment.reference_start, str(pileupread.alignment.query_name) + ";" + str(clip)], [pileupread.alignment.query_sequence[pileupread.query_position], pileupread.alignment.query_qualities[pileupread.query_position]])
total_n += 1
if deletions < (NanoSV.opts_max_deletions * total_n):
haplotypes = sorted(base_ratios.items(), key=lambda x: sum(x[1]))[-2:]
try:
if haplotypes[0][1][0] / sum(haplotypes[0][1]) > NanoSV.opts_min_occurences_of_highq_var and haplotypes[1][1][0] / sum(haplotypes[1][1]) > NanoSV.opts_min_occurences_of_highq_var:
if sum(haplotypes[0][1]) / (sum(haplotypes[1][1]) + sum(haplotypes[0][1])) > NanoSV.opts_min_occurences_of_var:
bin = int(int(snp_position) / NanoSV.opts_variant_bin_size)
variants[chromosome][bin][int(snp_position)] = variant
except ZeroDivisionError:
""
def find_variations_cigar(cigartuples, chromosome, pos, qual, seq, subsegments):
"""
Loops through read to find variant base. This function is used with LAST bam files
:param cigartuples:
:param chromosome:
:param pos:
:param qual:
:param seq:
:param subsegments:
"""
global tmp_variants
ref_cursor = (int(pos))
read_cursor = 0
for tuple in cigartuples:
if tuple[0] == 4:
read_cursor += tuple[1]
if tuple[0] == 8:
for mismatch in range(tuple[1]):
ref_cursor += 1
if ref_cursor not in tmp_variants:
tmp_variants[ref_cursor] = v.Variant(chromosome, ref_cursor)
for segment in subsegments:
if segment.pos <= ref_cursor <= segment.end:
tmp_variants[ref_cursor].add_segment(segment.id, [seq[read_cursor], qual[read_cursor]])
break
if segment.end < ref_cursor:
del subsegments[subsegments.index(segment)]
read_cursor += 1
elif tuple[0] == 2:
for deletion in range(tuple[1]):
ref_cursor += 1
if ref_cursor not in tmp_variants:
tmp_variants[ref_cursor] = v.Variant(chromosome, ref_cursor)
for segment in subsegments:
if segment.pos <= ref_cursor <= segment.end:
tmp_variants[ref_cursor].add_segment(segment.id, ['-'])
break
if segment.end < ref_cursor:
del subsegments[subsegments.index(segment)]
elif tuple[0] == 7:
ref_cursor += tuple[1]
read_cursor += tuple[1]
elif tuple[0] == 1:
read_cursor += tuple[1]