def frequent_words_with_mismatches_and_revcomps(text, k, d):
frequent_patterns = []
close = [0] * (4**k)
frequency_array = [0] * (4**k)
for i in range(0,len(text)-k+1):
neighborhood = neighbors(text[i:i+k], d)
for pattern in neighborhood:
index = pattern_to_number(pattern)
close[index] = 1
for i in range(0,4**k):
if close[i] == 1:
pattern = number_to_pattern(i, k)
frequency_array[i] = approximate_pattern_count(text, pattern, d)
frequency_array[i] += approximate_pattern_count(text, reverse_complement(pattern), d)
max_count = max(frequency_array)
for i in range(0,4**k):
if frequency_array[i] == max_count:
pattern = number_to_pattern(i, k)
if pattern not in frequent_patterns:
frequent_patterns.append(pattern)
return frequent_patterns
def count_frequency_array(sequence, klen):
frequency_array = [0] * (4**klen)
frequent_positions = []
start = 0
end = klen
for i in range(len(sequence) - klen):
frame = sequence[start:end]
num = pattern_to_number(frame)
frequency_array[num] = frequency_array[num] + 1
start = start + 1
end = end + 1
max_count = max(frequency_array)
for i in range(len(frequency_array)):
if frequency_array[i] == max_count:
frequent_positions.append(i)
# convert numbers to pattern
f = lambda x: number_to_pattern(x)
frequent_patterns = list(map(f, frequent_nums))
return frequency_array, frequent_positions, frequent_patterns
def better_clump_finding(genome, k, t, L):
frequent_patterns = []
clump = [0] * (4**k)
text = genome[0:L]
frequency_array = computing_frequencies(text,k)
for i in range(0,4**k):
if frequency_array[i] >= t:
clump[i] = 1
for i in range(1,len(genome)-L+1):
first_pattern = genome[i-1:i-1+k]
index = pattern_to_number(first_pattern)
frequency_array[index] = frequency_array[index] - 1
last_pattern = genome[i+L-k:i+L]
index = pattern_to_number(last_pattern)
frequency_array[index] = frequency_array[index] + 1
if frequency_array[index] >= t:
clump[index] = 1
for i in range(0,4**k):
if clump[i] == 1:
pattern = number_to_pattern(i,k)
if pattern not in frequent_patterns:
frequent_patterns.append(pattern)
return frequent_patterns
def prob_3010_4():
lines = open("data/dataset_3010_4.txt").read().splitlines()
number = int(lines[0])
k = int(lines[1])
fout = open("out.txt", "w")
fout.write(str(number_to_pattern(number, k)))
fout.close()
def clump_finding(genome, k, t, L):
frequent_patterns = []
clump = [0] * (4**k)
for i in range(0,len(genome)-L+1):
text = genome[i:i+L]
frequency_array = cf.computing_frequencies(text,k)
for index in range(0,4**k):
if frequency_array[index] >= t:
clump[index] = 1
for i in range(0,4**k):
if clump[i] == 1:
pattern = number_to_pattern(i, k)
frequent_patterns.append(pattern)
return frequent_patterns
