def get_approximate_frequent_words(text, k, d):
# Flag indicies of candidate kmers from neighborhoods
is_neighbor_flags = [0] * pow(4, k)
for position, kmer in ApproximatePatternMatching_1F.yield_kmers_by_position(text, k):
neighbors = calculate_neighbors(kmer, d)
for neighbor in neighbors:
number = frequency_array_lib.pattern_to_number(neighbor)
is_neighbor_flags[number] = 1
# Count approximate occurences of candidates
chunk_size = 5
translation_table = build_translation_table(chunk_size)
frequency_array = [0] * pow(4, k)
for i in range(len(is_neighbor_flags)):
if 1 == is_neighbor_flags[i]:
kmer = frequency_array_lib.number_to_pattern(i, k)
reverse_complement = generate_strand_complement(kmer, translation_table, chunk_size)
frequency = ApproximatePatternMatching_1F.count_approximate_matches(kmer, text, d) + ApproximatePatternMatching_1F.count_approximate_matches(reverse_complement, text, d)
frequency_array[i] = frequency
# Find max count
max_count = -1
for count in frequency_array:
if count > max_count:
max_count = count
# Find most frequent kmers
frequent = []
for i in range(len(frequency_array)):
if frequency_array[i] == max_count:
kmer = frequency_array_lib.number_to_pattern(i, k)
frequent.append(kmer)
return frequent
def compute_frequency_array(text, k):
fa = [0] * pow(4, k)
for kmer in frequency_array_lib.yield_kmers(text, k):
number = frequency_array_lib.pattern_to_number(kmer)
fa[number] = fa[number] + 1
return fa
def main():
with fileinput.input() as fi:
genome = fi.readline().rstrip()
number = frequency_array_lib.pattern_to_number(genome)
print(number)