def as_indel(self, ref_fasta):
chrom = self.chroms[0].lstrip('chr')
pos = self.breaks[0]
ref = alt = None
size = self.get_size()
if self.rearrangement == 'del':
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[1] - 1).upper()
alt = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[0]).upper()
else:
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[1]).upper()
alt = ref + self.novel_seq.upper()
id = self.id
qual = '.'
filter = '.'
info = {
'BKPTID': ','.join(self.contigs),
}
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
# repeat contraction
if self.rearrangement == 'del' and self.repeat_seq is not None:
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
if ref is not None and alt is not None:
fields = [
chrom, pos, id, ref, alt, qual, filter,
VCF.info_dict_to_str(info)
]
return '\t'.join(map(str, fields))
def as_indel(self, ref_fasta):
chrom = self.chroms[0].lstrip('chr')
pos = self.breaks[0]
ref = alt = None
size = self.get_size()
if self.rearrangement == 'del':
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[1] - 1).upper()
alt = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[0]).upper()
else:
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[1]).upper()
alt = ref + self.novel_seq.upper()
id = self.id
qual = '.'
filter = '.'
info = {
'BKPTID':','.join(self.contigs),
}
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
# repeat contraction
if self.rearrangement == 'del' and self.repeat_seq is not None:
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
if ref is not None and alt is not None:
fields = [chrom, pos, id, ref, alt, qual, filter, VCF.info_dict_to_str(info)]
return '\t'.join(map(str, fields))
def as_sv(self, ref_fasta, id_ext=None, info_ext=None, chrom_ext=None, pos_ext=None):
chrom = self.chroms[0] if chrom_ext is None else chrom_ext
pos = self.breaks[0] if pos_ext is None else pos_ext
chrom = chrom.lstrip('chr')
alt = None
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[0]).upper()
sv_len = self.get_size()
end = None
if type(sv_len) is int:
end = pos + sv_len
if self.rearrangement == 'del':
alt = '<DEL>'
sv_type = 'DEL'
if type(sv_len) is int:
sv_len = -1 * sv_len
end = pos - sv_len
elif self.rearrangement == 'dup':
alt = '<DUP:TANDEM>'
sv_type = 'DUP'
elif self.rearrangement == 'inv':
alt = '<INV>'
sv_type = 'INV'
elif self.rearrangement == 'ins':
alt = '<INS>'
sv_type = 'INS'
end = pos
id = self.id if id_ext is None else id_ext
qual = '.'
filter = '.'
info = {'SVTYPE': sv_type,
'END': end,
'BKPTID':','.join(self.contigs),
}
if end is not None:
info['END'] = end
if type(sv_len) is int:
info['SVLEN'] = sv_len
if sv_type == 'DUP':
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not None:
info['FLANKING_PAIRS'] = self.support['flanking']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-':
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
#print 'gmap', contig_breaks
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
cipos = '0,%d' % homol_len
if cipos is not None:
info['CIPOS'] = cipos
info['CIPOS'] = cipos
if homol_len is not None:
info['HOMLEN'] = homol_len
info['HOMLEN'] = homol_len
if homol_seq is not None:
info['HOMSEQ'] = homol_seq
info['HOMSEQ'] = homol_seq
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if key == 'SVLEN' and value == 'NA':
continue
info[key] = value
if ref is not None and alt is not None:
fields = [chrom, pos, id, ref, alt, qual, filter, VCF.info_dict_to_str(info)]
return '\t'.join(map(str, fields))
def as_breakends(self, ref_fasta, genomic=True, max_novel_seq_len=50, info_ext=None, parids=None, event=None):
chroms = map(lambda c: c.lstrip('chr'), self.chroms)
alt_chroms = chroms[:]
pos = list(self.breaks)
alt_pos = pos[:]
# inserted novel sequences
inserted_seqs = ['','']
if self.novel_seq and self.novel_seq != 'NA' and self.novel_seq != '-':
if len(self.novel_seq) > max_novel_seq_len:
alt_chroms[0] = '<%s>' % self.contigs[0]
alt_chroms[1] = '<%s>' % self.contigs[0]
alt_pos[1] = self.contig_breaks[0][0] + 1
alt_pos[0] = self.contig_breaks[0][1] - 1
else:
if len(self.aligns[0]) == 1:
inserted_seqs[0] = self.novel_seq if self.aligns[0][0].strand == '+' else reverse_complement(self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][0].strand == '+' else reverse_complement(self.novel_seq)
else:
inserted_seqs[0] = self.novel_seq if self.aligns[0][0].strand == '+' else reverse_complement(self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][1].strand == '+' else reverse_complement(self.novel_seq)
# microhomology, cipos
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-' and len(self.homol_seq) > 0:
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
pos[0] -= homol_len
alt_pos[1] += homol_len
cipos = '0,%d' % homol_len
refs = (ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1, self.breaks[0]).upper(),
ref_fasta.fetch(self.chroms[1], self.breaks[1] - 1, self.breaks[1]).upper())
ids = ('%s%s' % (self.id, 'a'),
'%s%s' % (self.id, 'b'))
svtype = 'BND' if genomic else 'FND'
infos = [{'SVTYPE':svtype, 'MATEID':ids[1], 'EVENTTYPE':self.rearrangement.upper()},
{'SVTYPE':svtype, 'MATEID':ids[0], 'EVENTTYPE':self.rearrangement.upper()}]
if cipos is not None:
infos[0]['CIPOS'] = cipos
infos[1]['CIPOS'] = cipos
if homol_len is not None:
infos[0]['HOMLEN'] = homol_len
infos[1]['HOMLEN'] = homol_len
if homol_seq is not None:
infos[0]['HOMSEQ'] = homol_seq
infos[1]['HOMSEQ'] = homol_seq
# read support
if self.final_support is not None:
#infos[0]['READSUPPORT'] = self.final_support
#infos[1]['READSUPPORT'] = self.final_support
infos[0]['SPANNING_READS'] = self.support['spanning']
infos[1]['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not NONE:
infos[0]['FLANKING_PAIRS'] = self.support['flanking']
infos[1]['FLANKING_PAIRS'] = self.support['flanking']
adj_size = self.get_size()
if type(adj_size) is int:
infos[0]['SVLEN'] = adj_size
infos[1]['SVLEN'] = adj_size
# somatic
if self.somatic:
infos[0]['SOMATIC'] = 'SOMATIC'
infos[1]['SOMATIC'] = 'SOMATIC'
# contig and contig breakpoints
if self.contigs:
for i in range(2):
infos[i]['BKPTID'] = ','.join(self.contigs)
if self.contig_breaks and len(self.contig_breaks) == len(self.contigs):
contig_breaks = []
for bk in self.contig_breaks:
if len(bk) == 2:
contig_breaks.append('%s-%s' % (bk[0], bk[1]))
else:
print 'error'
if len(contig_breaks) == len(self.contigs):
for i in range(2):
infos[i]['CTG_BKS'] = ','.join(contig_breaks)
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if len(value) == 2:
infos[0][key] = value[0]
infos[1][key] = value[1]
if self.orients[0] == 'L':
# LL
if self.orients[1] == 'L':
alts = ('%s%s]%s:%s]' % (refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
'%s%s]%s:%s]' % (refs[1], inserted_seqs[1], alt_chroms[0], alt_pos[0]))
# LR
else:
alts = ('%s%s[%s:%s[' % (refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
']%s:%s]%s%s' % (alt_chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
else:
# RL
if self.orients[1] == 'L':
alts = (']%s:%s]%s%s' % (chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'%s%s[%s:%s[' % (refs[1], inserted_seqs[1], chroms[0], alt_pos[0]))
# RR
else:
alts = ('[%s:%s[%s%s' % (chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'[%s:%s[%s%s' % (chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
breakends = map(lambda i: '\t'.join([chroms[i], str(pos[i]), ids[i], refs[i], alts[i], '.', '.', VCF.info_dict_to_str(infos[i])]), range(2))
return '\n'.join(breakends)
def as_sv(self,
ref_fasta,
id_ext=None,
info_ext=None,
chrom_ext=None,
pos_ext=None):
chrom = self.chroms[0] if chrom_ext is None else chrom_ext
pos = self.breaks[0] if pos_ext is None else pos_ext
chrom = chrom.lstrip('chr')
alt = None
ref = ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[0]).upper()
sv_len = self.get_size()
end = None
if type(sv_len) is int:
end = pos + sv_len
if self.rearrangement == 'del':
alt = '<DEL>'
sv_type = 'DEL'
if type(sv_len) is int:
sv_len = -1 * sv_len
end = pos - sv_len
elif self.rearrangement == 'dup':
alt = '<DUP:TANDEM>'
sv_type = 'DUP'
elif self.rearrangement == 'inv':
alt = '<INV>'
sv_type = 'INV'
elif self.rearrangement == 'ins':
alt = '<INS>'
sv_type = 'INS'
end = pos
id = self.id if id_ext is None else id_ext
qual = '.'
filter = '.'
info = {
'SVTYPE': sv_type,
'END': end,
'BKPTID': ','.join(self.contigs),
}
if end is not None:
info['END'] = end
if type(sv_len) is int:
info['SVLEN'] = sv_len
if sv_type == 'DUP':
if self.repeat_seq is not None:
info['REPEAT_SEQ'] = self.repeat_seq
if self.repeat_num is not None:
info['REPEAT_NUM'] = self.repeat_num
if self.repeat_num_change is not None:
info['REPEAT_NUM_CHANGE'] = self.repeat_num_change
# read support
if self.final_support is not None:
#info['READSUPPORT'] = self.final_support
info['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not None:
info['FLANKING_PAIRS'] = self.support['flanking']
# somatic
if self.somatic:
info['SOMATIC'] = 'SOMATIC'
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-':
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
#print 'gmap', contig_breaks
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
cipos = '0,%d' % homol_len
if cipos is not None:
info['CIPOS'] = cipos
info['CIPOS'] = cipos
if homol_len is not None:
info['HOMLEN'] = homol_len
info['HOMLEN'] = homol_len
if homol_seq is not None:
info['HOMSEQ'] = homol_seq
info['HOMSEQ'] = homol_seq
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if key == 'SVLEN' and value == 'NA':
continue
info[key] = value
if ref is not None and alt is not None:
fields = [
chrom, pos, id, ref, alt, qual, filter,
VCF.info_dict_to_str(info)
]
return '\t'.join(map(str, fields))
def as_breakends(self,
ref_fasta,
genomic=True,
max_novel_seq_len=50,
info_ext=None,
parids=None,
event=None):
chroms = map(lambda c: c.lstrip('chr'), self.chroms)
alt_chroms = chroms[:]
pos = list(self.breaks)
alt_pos = pos[:]
# inserted novel sequences
inserted_seqs = ['', '']
if self.novel_seq and self.novel_seq != 'NA' and self.novel_seq != '-':
if len(self.novel_seq) > max_novel_seq_len:
alt_chroms[0] = '<%s>' % self.contigs[0]
alt_chroms[1] = '<%s>' % self.contigs[0]
alt_pos[1] = self.contig_breaks[0][0] + 1
alt_pos[0] = self.contig_breaks[0][1] - 1
else:
if len(self.aligns[0]) == 1:
inserted_seqs[0] = self.novel_seq if self.aligns[0][
0].strand == '+' else reverse_complement(
self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][
0].strand == '+' else reverse_complement(
self.novel_seq)
else:
inserted_seqs[0] = self.novel_seq if self.aligns[0][
0].strand == '+' else reverse_complement(
self.novel_seq)
inserted_seqs[1] = self.novel_seq if self.aligns[0][
1].strand == '+' else reverse_complement(
self.novel_seq)
# microhomology, cipos
cipos = None
homol_len = None
homol_seq = None
if self.homol_seq and self.homol_seq[0] != '-' and len(
self.homol_seq) > 0:
homol_seq = self.homol_seq[0].upper()
homol_len = len(self.homol_seq[0])
contig_breaks = self.contig_breaks[0]
# e.g. GMAP
if contig_breaks[0] + 1 == contig_breaks[1]:
pass
# e.g. BWA-mem
elif contig_breaks[0] >= contig_breaks[1]:
pos[0] -= homol_len
alt_pos[1] += homol_len
cipos = '0,%d' % homol_len
refs = (ref_fasta.fetch(self.chroms[0], self.breaks[0] - 1,
self.breaks[0]).upper(),
ref_fasta.fetch(self.chroms[1], self.breaks[1] - 1,
self.breaks[1]).upper())
ids = ('%s%s' % (self.id, 'a'), '%s%s' % (self.id, 'b'))
svtype = 'BND' if genomic else 'FND'
infos = [{
'SVTYPE': svtype,
'MATEID': ids[1],
'EVENTTYPE': self.rearrangement.upper()
}, {
'SVTYPE': svtype,
'MATEID': ids[0],
'EVENTTYPE': self.rearrangement.upper()
}]
if cipos is not None:
infos[0]['CIPOS'] = cipos
infos[1]['CIPOS'] = cipos
if homol_len is not None:
infos[0]['HOMLEN'] = homol_len
infos[1]['HOMLEN'] = homol_len
if homol_seq is not None:
infos[0]['HOMSEQ'] = homol_seq
infos[1]['HOMSEQ'] = homol_seq
# read support
if self.final_support is not None:
#infos[0]['READSUPPORT'] = self.final_support
#infos[1]['READSUPPORT'] = self.final_support
infos[0]['SPANNING_READS'] = self.support['spanning']
infos[1]['SPANNING_READS'] = self.support['spanning']
if self.support['flanking'] is not NONE:
infos[0]['FLANKING_PAIRS'] = self.support['flanking']
infos[1]['FLANKING_PAIRS'] = self.support['flanking']
adj_size = self.get_size()
if type(adj_size) is int:
infos[0]['SVLEN'] = adj_size
infos[1]['SVLEN'] = adj_size
# somatic
if self.somatic:
infos[0]['SOMATIC'] = 'SOMATIC'
infos[1]['SOMATIC'] = 'SOMATIC'
# contig and contig breakpoints
if self.contigs:
for i in range(2):
infos[i]['BKPTID'] = ','.join(self.contigs)
if self.contig_breaks and len(self.contig_breaks) == len(self.contigs):
contig_breaks = []
for bk in self.contig_breaks:
if len(bk) == 2:
contig_breaks.append('%s-%s' % (bk[0], bk[1]))
else:
print 'error'
if len(contig_breaks) == len(self.contigs):
for i in range(2):
infos[i]['CTG_BKS'] = ','.join(contig_breaks)
# external info - overrides given info
if info_ext:
for key, value in info_ext.iteritems():
if len(value) == 2:
infos[0][key] = value[0]
infos[1][key] = value[1]
if self.orients[0] == 'L':
# LL
if self.orients[1] == 'L':
alts = ('%s%s]%s:%s]' %
(refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
'%s%s]%s:%s]' %
(refs[1], inserted_seqs[1], alt_chroms[0], alt_pos[0]))
# LR
else:
alts = ('%s%s[%s:%s[' %
(refs[0], inserted_seqs[0], alt_chroms[1], alt_pos[1]),
']%s:%s]%s%s' %
(alt_chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
else:
# RL
if self.orients[1] == 'L':
alts = (']%s:%s]%s%s' %
(chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'%s%s[%s:%s[' %
(refs[1], inserted_seqs[1], chroms[0], alt_pos[0]))
# RR
else:
alts = ('[%s:%s[%s%s' %
(chroms[1], alt_pos[1], inserted_seqs[0], refs[0]),
'[%s:%s[%s%s' %
(chroms[0], alt_pos[0], inserted_seqs[1], refs[1]))
breakends = map(
lambda i: '\t'.join([
chroms[i],
str(pos[i]), ids[i], refs[i], alts[i], '.', '.',
VCF.info_dict_to_str(infos[i])
]), range(2))
return '\n'.join(breakends)