if __name__ == "__main__":
mismatches = 1
for record in sys.stdin:
index = None
for col in record.strip().split():
col = col.upper()
if len([c for c in col if c not in "ACGTN-"]) > 0:
continue
index = col
break
if index is None:
continue
names = []
for name, sequence in BASIC_LOOKUP.items():
try:
dist = hamming_distance(index, sequence)
if dist <= mismatches:
names.append([name, dist])
except:
pass
print(
"\t".join(
record.strip().split()
+ [",".join(sorted([n[0] for n in names if n[1] == i])) for i in range(mismatches + 1)]
)
)
outfile = csv.writer(open('index_overlaps.csv', 'wb'))
outfile.writerow(['Index', 'Sequence', 'Identical',
'One mismatch', 'Two mismatches'])
def compare(key1, key2):
for pos in xrange(len(key1)-1, -1, -1):
if not key1[pos].isdigit():
k1 = (key1[0:pos+1], int(key1[pos+1:]))
break
for pos in xrange(len(key2)-1, -1, -1):
if not key2[pos].isdigit():
k2 = (key2[0:pos+1], int(key2[pos+1:]))
break
return cmp(k1, k2)
keys = sorted(BASIC_LOOKUP.keys(), cmp=compare)
for key1 in keys:
seq1 = BASIC_LOOKUP[key1]
identical = []
one = []
two = []
for key2 in keys:
if key1 == key2: continue
seq2 = BASIC_LOOKUP[key2]
mismatch = 0
for n1, n2 in zip(seq1, seq2):
if n1 != n2: mismatch += 1
if mismatch == 0:
identical.append(key2)
elif mismatch == 1: