def frequent_words_with_mismatches(dna, k, d):
frequent_patterns = []
close_array = []
frequency_array = []
for i in range(4**k):
close_array.append(0)
frequency_array.append(0)
for i in range(len(dna) - k + 1):
neighbourhood = neighbours(dna[i:i + k], d)
for pattern in neighbourhood:
idx = PatternToNumber(pattern)
close_array[idx] = 1
for i in range(4**k):
if close_array[i] == 1:
pattern = NumberToPattern(i, k)
frequency_array[i] = len(
approximate_pattern_matching(pattern, dna, d))
max_value = max(frequency_array)
for i in range(4**k):
if frequency_array[i] == max_value:
pattern = NumberToPattern(i, k)
frequent_patterns.append(pattern)
return frequent_patterns
def clump_finding(dna, k, L, t):
clump_frequent_pattern_list = []
clump_array = []
for i in range(4**k - 1):
clump_array.append(0)
frequency_array = computing_frequencies(dna[0:L], k)
for i in range(4**k - 1):
if (frequency_array[i] >= t):
pattern = NumberToPattern(i, k)
clump_frequent_pattern_list.append(pattern)
for i in range(1, len(dna) - L + 1):
first_pattern = dna[i - 1:i - 1 + k]
index = PatternToNumber(first_pattern)
frequency_array[index] = frequency_array[index] - 1
second_pattern = dna[i + L - k:i + L]
index = PatternToNumber(second_pattern)
frequency_array[index] = frequency_array[index] + 1
if (frequency_array[index] >= t):
clump_array[index] = 1
for i in range(4**k - 1):
if (clump_array[i] == 1):
pattern = NumberToPattern(i, k)
clump_frequent_pattern_list.append(pattern)
return clump_frequent_pattern_list
def finding_frequent_patterns_sorting(dna, k):
frequent_patterns = []
index = []
count = []
for i in range(len(dna) - k + 1):
pattern = dna[i:i + k]
idx = PatternToNumber(pattern)
cnt = 1
index.append(idx)
count.append(cnt)
index.sort()
for i in range(1, len(dna) - k + 1):
if (index[i] == index[i - 1]):
count[i] = count[i - 1] + 1
max_count = max(count)
for i in range(0, len(dna) - k + 1):
if (count[i] == max_count):
pattern = NumberToPattern(index[i], k)
frequent_patterns.append(pattern)
return frequent_patterns
def profile_kmer(dna, k, profile):
pattern_list=[]
pattern_prob=[]
for i in range(4**k):
kmer = NumberToPattern(i,k)
pattern_list.append(kmer)
prob = pattern_probability(kmer, profile)
pattern_prob.append(prob)
pattern_prob, pattern_list = (list(t) for t in zip(*sorted(zip(pattern_prob, pattern_list))))
for i in range(len(pattern_prob)-1,0,-1):
for j in range(len(dna)-k+1):
if(dna[j:j+k]==pattern_list[i]):
return dna[j:j+k]
def median_string(dna_list, k):
distance = math.inf
median = ''
pattern_list = []
for i in range(4**k):
kmer = NumberToPattern(i,k)
pattern_list.append(kmer)
for pattern in pattern_list:
score=0
for seq in dna_list:
score = score + min_hamming_distance(pattern, seq)
if score < distance:
distance = score
median = pattern
return median