for i in xrange(len(data)-(k-1)):
d[s[i:i+k]] += 1
return d
def permutation_test(s, n=100):
"""Assess odds ratio on permuted string and return a list with scores."""
res = []
for _ in range(n):
random.shuffle(s)
f = count_kmers("".join(s), 2)
res.append(log2(N * f[a + b] / float(f1[a] * f1[b])))
return res
from main import get_seq
data = "".join(get_seq())
N = float(len(data))
f2 = count_kmers(data, 2)
f1 = count_kmers(data, 1)
a = "G"; b = "C"
odds_ratio = log2(N * f2[a + b] / float(f1[a] * f1[b]))
print odds_ratio
list_data = list(data)
res = permutation_test(list_data)
p = sum(1 for r in res if r>odds_ratio) / float(len(res))
print "OR=%5.3f, p=%7.5f" % (odds_ratio, p)
offset = 0
skipped = 0
for i in xrange(min(len(seq1), len(seq2))):
if i + offset == len(seq2)-1:
break
if(seq1[i] != seq2[i+offset]):
if(seq1[i] == seq2[i+offset+1] and seq1[i+1] == seq2[i+offset+2]):
offset += 1
skipped += 1
continue
diff+=1
print "skipped ", skipped
return diff+skipped
if __name__ == '__main__':
seqs = [get_seq("NC_001807.1.fasta"), get_seq("NC_001807.2.fasta"), get_seq("NC_001807.3.fasta"),
get_seq("NC_001807.4.fasta"), get_seq("NC_012920.1.fasta")]
lens = []
for s in seqs[:-1]:
lens.append(simple_seq_diff(seqs[4], s))
print ""
import pylab
pylab.bar(range(len(lens)), lens)
pylab.xlabel("NC_001807.1 NC_001807.2 NC_001807.3 NC_001807.4")
pylab.ylabel("Num differences + num skips")
pylab.show()
