def check_align(self, kread, mer, skmers, type='setup'):
match = False
v1 = olc.nw(self.aseq.seq, kread.seq)
v2 = olc.nw(kread.seq, self.aseq.seq)
# print 'check_align()'
# print 'Consensus seq', self.aseq.seq
# print 'Read seq', kread.seq
# print v1
# print v2
min_score = float(min(len(self.aseq.seq), len(kread.seq))) / 4.0
ident1 = float(v1[6]) / float(v1[2] - v1[3])
ident2 = float(v2[6]) / float(v2[2] - v2[3])
# print 'Min score', min_score, ', identity of overlap segments', ident1, ident2
if (v1[6] < min_score or ident1 < 0.9) and (v2[6] < min_score
or ident2 < 0.9):
return False
if v1[6] == v2[6] and v1[3] == 0 and v1[5] == 0:
# print 'Consensus and read sequence are the same'
return True
if v1[6] == v2[6]:
match = True
if len(self.aseq.seq) < len(kread.seq) or (v1[2] == len(
self.aseq.seq) and v1[3] == 0):
# consensus sequence is a subseq of kread
self.aseq = assembly_seq(kread.seq)
# print 'Consensus is subseq of kread'
self.aseq.add_subseq(v1[5], v1[4])
if type == 'grow':
self.set_kmers(skmers)
elif len(kread.seq) < len(self.aseq.seq) or (v2[2] == len(
kread.seq) and v2[3] == 0):
# print 'Consensus contains read seq'
self.aseq.add_subseq(v2[5], v2[4])
else:
match = False
indx11 = v1[0].replace('-', '').find(mer)
indx12 = v1[1].replace('-', '').find(mer)
indx21 = v2[0].replace('-', '').find(mer)
indx22 = v2[1].replace('-', '').find(mer)
if indx11 > -1 and indx12 > -1:
if (indx21 == -1 and indx22 == -1) or (
abs(indx21 - indx22) > abs(indx11 - indx12)):
match = True
self.contig_overlap_read(v1, kread, skmers, type)
elif indx21 > -1 and indx22 > -1:
if (indx11 == -1 and indx12 == -1) or (
abs(indx21 - indx22) < abs(indx11 - indx12)):
match = True
self.read_overlap_contig(v2, kread, skmers, type)
elif v1[6] > v2[6]:
match = True
self.contig_overlap_read(v1, kread, skmers, type)
else:
# Read hangs left
match = True
self.read_overlap_contig(v2, kread, skmers, type)
return match
def check_align(self, kread, mer, nreads, skmers, type='setup') :
match = False
v1 = olc.nw(self.aseq.seq, kread.seq)
v2 = olc.nw(kread.seq, self.aseq.seq)
# print 'check_align()'
# print 'Consensus seq', self.aseq.seq
# print 'Read seq', kread.seq
# print 'Resolve counts' , resolve_counts
# print v1
# print v2
min_score = float(min(len(self.aseq.seq), len(kread.seq) )) / 4.0
ident1 = round(float(v1[6]) / float(v1[2] - v1[3]), 2)
ident2 = round(float(v2[6]) / float(v2[2] - v2[3]), 2)
# print 'Min score', min_score, ', identity of overlap segments', ident1, ident2
if (v1[6] < min_score or ident1 < 0.90) and (v2[6] < min_score or ident2 < 0.90) :
return False
if v1[6] == v2[6] and v1[3] == 0 and v1[5] == 0 and len(self.aseq.seq) == len(kread.seq) :
# print 'Consensus and read sequence are the same'
return True
if v1[6] == v2[6] :
match = True
if len(self.aseq.seq) < len(kread.seq) or (v1[2] == len(self.aseq.seq) and v1[3] == 0) :
# consensus sequence is a subseq of kread
# max_counts = max(self.aseq.counts)
# self.aseq = assembly_seq(kread, nreads)
# print 'Consensus is subseq of kread'
self.aseq.set_superseq(kread, nreads, v1[5], v1[4])
# self.aseq.add_subseq(v1[5], v1[4], max_counts, kread.indel_only)
if type == 'grow' :
self.set_kmers(skmers)
elif len(kread.seq) < len(self.aseq.seq) or (v2[2] == len(kread.seq) and v2[3] == 0):
# print 'Consensus contains read seq'
self.aseq.add_subseq(v2[5], v2[4], nreads, kread.indel_only)
else :
match = False
indx11 = v1[0].replace('-','').find(mer)
indx12 = v1[1].replace('-','').find(mer)
indx21 = v2[0].replace('-','').find(mer)
indx22 = v2[1].replace('-','').find(mer)
if indx11 > -1 and indx12 > -1 :
if (indx21 == -1 and indx22 == -1) or (abs(indx21 - indx22) > abs(indx11 - indx12)) :
match = True
self.contig_overlap_read(v1, kread, nreads, skmers, type)
elif indx21 > -1 and indx22 > -1 :
if (indx11 == -1 and indx12 == -1) or (abs(indx21 - indx22) < abs(indx11 - indx12)) :
match = True
self.read_overlap_contig(v2, kread, nreads, skmers, type)
elif v1[6] > v2[6] :
match = True
self.contig_overlap_read(v1, kread, nreads, skmers, type)
else :
# Read hangs left
match = True
self.read_overlap_contig(v2, kread, nreads, skmers, type)
return match
def subseq(seq1, seq2) :
aln = olc.nw(seq2, seq1)
seq2_sub = (False,None)
if aln[2] == len(seq2) and aln[3] == 0 and aln[6] >= (0.90*(len(seq2))) :
if len(seq2) < len(seq1) :
seq2_sub = (True,None)
else :
seq2_sub = (True,aln[6])
else :
seq2_sub = (False,aln[6])
return seq2_sub
def subseq(seq1, seq2):
aln = olc.nw(seq2, seq1)
seq2_sub = (False, None)
if aln[2] == len(seq2) and aln[3] == 0 and aln[6] >= (0.90 * (len(seq2))):
if len(seq2) < len(seq1):
seq2_sub = (True, None)
else:
seq2_sub = (True, aln[6])
else:
seq2_sub = (False, aln[6])
return seq2_sub
def same_reads(seq1, seq2) :
same = False
aln = olc.nw(seq1, seq2)
if aln[3] == 0 and aln[5] == 0 and aln[6] > 0.95*(len(seq1)) :
same = True
return same
def same_reads(seq1, seq2):
same = False
aln = olc.nw(seq1, seq2)
if aln[3] == 0 and aln[5] == 0 and aln[6] > 0.95 * (len(seq1)):
same = True
return same