def wrapperRMS(dna, k, t, iter):
motifs = randomizedMotifSearch(dna, k, t)
for i in range(iter - 1):
newMotifs = randomizedMotifSearch(dna, k, t)
if score(newMotifs) < score(motifs):
motifs = newMotifs
return motifs
def wrapperGS(dna, k, t, n, iter):
motifs = gibbsSampler(dna, k, t, n)
for i in range(iter - 1):
newMotifs = gibbsSampler(dna, k, t, n)
if score(newMotifs) < score(motifs):
motifs = newMotifs
return motifs
def randomizedMotifSearch(dna, k, t):
bestMotifs = randomMotifs(dna, k, t)
while True:
profile = profileMotifs(bestMotifs)
newMotifs = motifs(profile, dna)
if score(newMotifs) < score(bestMotifs):
bestMotifs = newMotifs
else:
return bestMotifs
def greedyMotifSearch(dna, k, t):
#computes a collection of strings as best motifs.
bestMotifs = [row[:k] for row in dna[:t]]
for i in range(len(dna[0]) - k + 1):
motif1 = dna[0][i:i + k]
candidateMotifs = [motif1]
for j in range(1, t):
profile = profileMotifs(candidateMotifs)
motif_i = profileMostProbableKmer(dna[j], k, profile)
candidateMotifs.append(motif_i)
if (score(candidateMotifs) < score(bestMotifs)):
bestMotifs = candidateMotifs
return bestMotifs
def gibbsSampler(dna, k, t, n):
# applys Gibss sampling
bestMotifs = randomMotifs(dna, k, t)
for i in range(n):
index = randint(0, len(dna) - 1)
newMotifs = bestMotifs
newMotifs.pop(index)
profile = profileMotifs(newMotifs)
probDistr = probDistribution(dna[index], profile, k)
indexBiased = randomWithBias(probDistr)
newMotifs.insert(index, dna[index][indexBiased:indexBiased + k])
if score(newMotifs) < score(bestMotifs):
bestMotifs = newMotifs
return bestMotifs
def randomizedMotifSearch(dna, k, t):
bestMotifs = randomMotifs(dna, k, t)
while True:
profile = profileMotifs(bestMotifs)
newMotifs = motifs(profile, dna)
if score(newMotifs) < score(bestMotifs):
bestMotifs = newMotifs
else:
return bestMotifs
def wrapperRMS(dna, k, t, iter):
motifs = randomizedMotifSearch(dna, k, t)
for i in range(iter - 1):
newMotifs = randomizedMotifSearch(dna, k, t)
if score(newMotifs) < score(motifs):
motifs = newMotifs
return motifs
if __name__ == "__main__":
file = open("data/subtle_motif_dataset.txt", 'r')
dna = [line.rstrip('\n') for line in file]
result_motifs = wrapperRMS(dna, 15, 10, 1000)
score_motifs = score(result_motifs)
consensusMotifs = consensus(result_motifs)
verticalPrint(result_motifs)
print(score_motifs)
print(consensusMotifs)