def k_mer_generator(k):
"""returns a set of all 4^k k-mers"""
k_mer_list = []
for i in range(4**k - 1):
k_mer_list.append(ntp.number_to_pattern(i, k))
return set(k_mer_list)
def freq_seq_mismatch_sort(genome, k, d):
frequent_patterns = set()
neighborhoods = []
count = [0]
pos_index = [0]
for i in range(len(genome) - k):
neighborhoods.append(neighbors.recursive_neighbors(genome[i:i + k], d))
flattened_neighborhood = [i for i in apc.iter_flatten(neighborhoods)]
for i in range(len(flattened_neighborhood)):
pattern = flattened_neighborhood[i]
pos_index.append(ptn.pattern_to_number(pattern))
count.append(1)
sorted_index = sorted(pos_index)
for i in range(len(flattened_neighborhood)):
if sorted_index[i] == sorted_index[i + 1]:
count[i + 1] = count[i] + 1
max_count = max(count)
for i in range(len(flattened_neighborhood)):
if count[i] == max_count:
pattern = ntp.number_to_pattern(sorted_index[i], k)
frequent_patterns.add(pattern)
return frequent_patterns
def all_kmer_generator(KMER_LENGTH):
""" Generates a set of all possible kmers given the input KMER_LENGTH """
kmer_set = set()
for i in range(4 ** KMER_LENGTH - 1):
kmer_set.add(ntp.number_to_pattern(i, KMER_LENGTH))
return kmer_set
def freq_words_w_mismatches(text, k, d):
freq_patterns = set()
freq_patterns_w_reversecomp = []
close_array = [0]
freq_array = [0]
neighborhood = []
for i in range(4**k - 1):
close_array.append(0)
freq_array.append(0)
for i in range(len(text) - k):
neighborhood.append(neighbors.recursive_neighbors(text[i:i+k], d))
flat_neighborhood = [i for i in apc.iter_flatten(neighborhood)]
for pattern in flat_neighborhood:
the_index = ptn.pattern_to_number(pattern)
close_array[the_index] = 1
for i in range(4**k - 1):
if close_array[i] == 1:
generated_pattern = ntp.number_to_pattern(i, k)
freq_array[i] = apc.approx_pattern_count(text, generated_pattern, d)
rc_pattern = rc.reverse_complement(generated_pattern)
freq_patterns_w_reversecomp.append(rc_pattern)
the_max_count = max(freq_array)
for i in range(4**k - 1):
if freq_array[i] == the_max_count:
generated_kmer = ntp.number_to_pattern(i, k)
freq_patterns.add(generated_kmer)
return freq_patterns
def median_string(dna_list, k):
"""generates the median string that has the minimum distance"""
median_list = []
distance = k * (len(dna_list) - 1)
for i in range(4**k - 1):
k_mer = ntp.number_to_pattern(i, k)
distance_pattern = distance_pattern_string(k_mer, dna_list)
if distance > distance_pattern:
distance = distance_pattern
median = k_mer
median_list = []
if distance == distance_pattern:
median_list.append(k_mer)
return median_list