def ctgs_overlap(ctg1, ctg2):
head_seq1 = ctg_seq[ctg1][:100]
tail_seq1 = ctg_seq[ctg1][-100:]
head_seq2 = ctg_seq[ctg2][:100]
tail_seq2 = ctg_seq[ctg2][-100:]
hh_align = myalign(head_seq1, reverse_complement(head_seq2))
hh_score = hh_align[2]
ht_align = myalign(head_seq1, tail_seq2)
ht_score = ht_align[2]
th_align = myalign(tail_seq1, head_seq2)
th_score = th_align[2]
tt_align = myalign(tail_seq1, reverse_complement(tail_seq2))
tt_score = tt_align[2]
# hh_align = nw.global_align(head_seq1, reverse_complement(head_seq2))
# hh_score = len([x for x,y in zip(*hh_align) if x == y])
# ht_align = nw.global_align(head_seq1, tail_seq2)
# ht_score = len([x for x,y in zip(*ht_align) if x == y])
# th_align = nw.global_align(tail_seq1, head_seq2)
# th_score = len([x for x,y in zip(*th_align) if x == y])
# tt_align = nw.global_align(tail_seq1, reverse_complement(tail_seq2))
# tt_score = len([x for x,y in zip(*tt_align) if x == y])
scores = (hh_score, ht_score, th_score, tt_score)
aligns = (hh_align, ht_align, th_align, tt_align)
best_i = max([0,1,2,3],key=lambda x: scores[x])
# print best_i
# print scores[best_i]
# print aligns[best_i][0]
# print aligns[best_i][1]
if scores[best_i] > 100:
return True
def ctgs_overlap(ctg1, ctg2):
head_seq1 = ctg_seq[ctg1][:100]
tail_seq1 = ctg_seq[ctg1][-100:]
head_seq2 = ctg_seq[ctg2][:100]
tail_seq2 = ctg_seq[ctg2][-100:]
hh_align = myalign(head_seq1, reverse_complement(head_seq2))
hh_score = hh_align[2]
ht_align = myalign(head_seq1, tail_seq2)
ht_score = ht_align[2]
th_align = myalign(tail_seq1, head_seq2)
th_score = th_align[2]
tt_align = myalign(tail_seq1, reverse_complement(tail_seq2))
tt_score = tt_align[2]
# hh_align = nw.global_align(head_seq1, reverse_complement(head_seq2))
# hh_score = len([x for x,y in zip(*hh_align) if x == y])
# ht_align = nw.global_align(head_seq1, tail_seq2)
# ht_score = len([x for x,y in zip(*ht_align) if x == y])
# th_align = nw.global_align(tail_seq1, head_seq2)
# th_score = len([x for x,y in zip(*th_align) if x == y])
# tt_align = nw.global_align(tail_seq1, reverse_complement(tail_seq2))
# tt_score = len([x for x,y in zip(*tt_align) if x == y])
scores = (hh_score, ht_score, th_score, tt_score)
aligns = (hh_align, ht_align, th_align, tt_align)
best_i = max([0, 1, 2, 3], key=lambda x: scores[x])
# print best_i
# print scores[best_i]
# print aligns[best_i][0]
# print aligns[best_i][1]
if scores[best_i] > 100:
return True
line1 = line1[1:-1]
if ':' in line1:
# we have an edge
origin, neighbors = line1.split(':')
if origin.endswith("'"):
origin_v = origin[:-1]
origin_o = 1
else:
origin_v = origin
origin_o = 0
# add origin as vertex
if origin_o == 0:
v_seq = line2
else:
v_seq = reverse_complement(line2)
add_vertex(origin_v, v_seq)
for neighbor in neighbors.split(','):
if neighbor.endswith("'"):
edges.append( (origin_v, origin_o, neighbor[:-1], 1) )
else:
edges.append( (origin_v, origin_o, neighbor, 0) )
else:
# we have a vertex
if line1.endswith("'"):
v_name = line1[:-1]
ori = 1
else:
v_name = line1
ori = 0
line1 = line1[1:-1]
if ':' in line1:
# we have an edge
origin, neighbors = line1.split(':')
if origin.endswith("'"):
origin_v = origin[:-1]
origin_o = 1
else:
origin_v = origin
origin_o = 0
# add origin as vertex
if origin_o == 0:
v_seq = line2
else:
v_seq = reverse_complement(line2)
add_vertex(origin_v, v_seq)
for neighbor in neighbors.split(','):
if neighbor.endswith("'"):
edges.append((origin_v, origin_o, neighbor[:-1], 1))
else:
edges.append((origin_v, origin_o, neighbor, 0))
else:
# we have a vertex
if line1.endswith("'"):
v_name = line1[:-1]
ori = 1
else:
v_name = line1
ori = 0