def greedy_motif_search(dna, k, t, pseudo=0):
"""
:param dna: DNA strings
:param k: kmer size
:param t: number of kmers in the DNA
:param pseudo: pseudocount
:return: best motif
"""
n = len(dna[0])
best_motifs = []
best_score = 0
for i in range(n - k + 1):
motifs = [dna[0][i:i + k]]
for j in range(1, t):
profile = get_profile(get_count(motifs[0:j]))
if pseudo:
profile = add_scalar_to_matrix(pseudo, profile)
motifs.append(profile_most_probable_pattern(dna[j], k, profile))
current_score = score(motifs, consensus(get_count(motifs)))
if not best_motifs or current_score < best_score:
best_motifs = motifs
best_score = current_score
return best_motifs
def test_greedy_motif_search_heavy(self):
dna = [
'GCGCCCCGCCCGGACAGCCATGCGCTAACCCTGGCTTCGATGGCGCCGGCTCAGTTAGGGCCGGAAGTCCCCAATGTGGCAGACCTTTCGCCCCTGGCGGACGAATGACCCCAGTGGCCGGGACTTCAGGCCCTATCGGAGGGCTCCGGCGCGGTGGTCGGATTTGTCTGTGGAGGTTACACCCCAATCGCAAGGATGCATTATGACCAGCGAGCTGAGCCTGGTCGCCACTGGAAAGGGGAGCAACATC',
'CCGATCGGCATCACTATCGGTCCTGCGGCCGCCCATAGCGCTATATCCGGCTGGTGAAATCAATTGACAACCTTCGACTTTGAGGTGGCCTACGGCGAGGACAAGCCAGGCAAGCCAGCTGCCTCAACGCGCGCCAGTACGGGTCCATCGACCCGCGGCCCACGGGTCAAACGACCCTAGTGTTCGCTACGACGTGGTCGTACCTTCGGCAGCAGATCAGCAATAGCACCCCGACTCGAGGAGGATCCCG',
'ACCGTCGATGTGCCCGGTCGCGCCGCGTCCACCTCGGTCATCGACCCCACGATGAGGACGCCATCGGCCGCGACCAAGCCCCGTGAAACTCTGACGGCGTGCTGGCCGGGCTGCGGCACCTGATCACCTTAGGGCACTTGGGCCACCACAACGGGCCGCCGGTCTCGACAGTGGCCACCACCACACAGGTGACTTCCGGCGGGACGTAAGTCCCTAACGCGTCGTTCCGCACGCGGTTAGCTTTGCTGCC',
'GGGTCAGGTATATTTATCGCACACTTGGGCACATGACACACAAGCGCCAGAATCCCGGACCGAACCGAGCACCGTGGGTGGGCAGCCTCCATACAGCGATGACCTGATCGATCATCGGCCAGGGCGCCGGGCTTCCAACCGTGGCCGTCTCAGTACCCAGCCTCATTGACCCTTCGACGCATCCACTGCGCGTAAGTCGGCTCAACCCTTTCAAACCGCTGGATTACCGACCGCAGAAAGGGGGCAGGAC',
'GTAGGTCAAACCGGGTGTACATACCCGCTCAATCGCCCAGCACTTCGGGCAGATCACCGGGTTTCCCCGGTATCACCAATACTGCCACCAAACACAGCAGGCGGGAAGGGGCGAAAGTCCCTTATCCGACAATAAAACTTCGCTTGTTCGACGCCCGGTTCACCCGATATGCACGGCGCCCAGCCATTCGTGACCGACGTCCCCAGCCCCAAGGCCGAACGACCCTAGGAGCCACGAGCAATTCACAGCG',
'CCGCTGGCGACGCTGTTCGCCGGCAGCGTGCGTGACGACTTCGAGCTGCCCGACTACACCTGGTGACCACCGCCGACGGGCACCTCTCCGCCAGGTAGGCACGGTTTGTCGCCGGCAATGTGACCTTTGGGCGCGGTCTTGAGGACCTTCGGCCCCACCCACGAGGCCGCCGCCGGCCGATCGTATGACGTGCAATGTACGCCATAGGGTGCGTGTTACGGCGATTACCTGAAGGCGGCGGTGGTCCGGA',
'GGCCAACTGCACCGCGCTCTTGATGACATCGGTGGTCACCATGGTGTCCGGCATGATCAACCTCCGCTGTTCGATATCACCCCGATCTTTCTGAACGGCGGTTGGCAGACAACAGGGTCAATGGTCCCCAAGTGGATCACCGACGGGCGCGGACAAATGGCCCGCGCTTCGGGGACTTCTGTCCCTAGCCCTGGCCACGATGGGCTGGTCGGATCAAAGGCATCCGTTTCCATCGATTAGGAGGCATCAA',
'GTACATGTCCAGAGCGAGCCTCAGCTTCTGCGCAGCGACGGAAACTGCCACACTCAAAGCCTACTGGGCGCACGTGTGGCAACGAGTCGATCCACACGAAATGCCGCCGTTGGGCCGCGGACTAGCCGAATTTTCCGGGTGGTGACACAGCCCACATTTGGCATGGGACTTTCGGCCCTGTCCGCGTCCGTGTCGGCCAGACAAGCTTTGGGCATTGGCCACAATCGGGCCACAATCGAAAGCCGAGCAG',
'GGCAGCTGTCGGCAACTGTAAGCCATTTCTGGGACTTTGCTGTGAAAAGCTGGGCGATGGTTGTGGACCTGGACGAGCCACCCGTGCGATAGGTGAGATTCATTCTCGCCCTGACGGGTTGCGTCTGTCATCGGTCGATAAGGACTAACGGCCCTCAGGTGGGGACCAACGCCCCTGGGAGATAGCGGTCCCCGCCAGTAACGTACCGCTGAACCGACGGGATGTATCCGCCCCAGCGAAGGAGACGGCG',
'TCAGCACCATGACCGCCTGGCCACCAATCGCCCGTAACAAGCGGGACGTCCGCGACGACGCGTGCGCTAGCGCCGTGGCGGTGACAACGACCAGATATGGTCCGAGCACGCGGGCGAACCTCGTGTTCTGGCCTCGGCCAGTTGTGTAGAGCTCATCGCTGTCATCGAGCGATATCCGACCACTGATCCAAGTCGGGGGCTCTGGGGACCGAAGTCCCCGGGCTCGGAGCTATCGGACCTCACGATCACC',
]
k = 15
t = len(dna)
expected = [
'GTTAGGGCCGGAAGT', 'CCGATCGGCATCACT', 'ACCGTCGATGTGCCC',
'GGGTCAGGTATATTT', 'GTGACCGACGTCCCC', 'CTGTTCGCCGGCAGC',
'CTGTTCGATATCACC', 'GTACATGTCCAGAGC', 'GCGATAGGTGAGATT',
'CTCATCGCTGTCATC'
]
from motifs.count_matrix import get_count
expected_consensus = 'gttAAAtAgaGatGtG'.upper()
expected_score = 58
result = greedy_motif_search(dna, k, t)
self.assertEqual(expected, result)
cons = consensus(get_count(result))
print(cons)
def gibbs_sampler(dna, k, t, n):
best_score = float('inf')
motifs = random_motifs([string for string in dna], k, t)
best_motifs = motifs
for j in range(n):
skip_index = randint(0, t - 1)
dna_slice = dna[:skip_index] + dna[skip_index + 1:]
profile = profile_with_pseudocounts(get_count(dna_slice))
motifs[skip_index] = profile_generated_string(motifs[skip_index],
profile, k)
current_score = score(motifs, consensus(get_count(motifs)))
if current_score < best_score:
best_motifs = motifs
best_score = current_score
return best_motifs
def repeated_gibbs_sampler(dna, k, t, n):
best_score = float('inf')
best_motifs = []
for i in range(100):
motifs = gibbs_sampler(dna, k, t, n)
current_score = score(motifs, consensus(get_count(motifs)))
if current_score < best_score:
best_score = current_score
best_motifs = motifs
return best_motifs
def test_count_with_pseudocounts(self):
expected = {
'A': [3, 4, 4, 4, 7, 5, 3, 3, 2, 4],
'C': [3, 4, 5, 4, 3, 4, 3, 2, 4, 4],
'T': [3, 3, 1, 5, 2, 1, 3, 3, 2, 5],
'G': [5, 3, 4, 1, 2, 4, 5, 6, 6, 1]
}
count = get_count(self.motifs)
result = add_scalar_to_matrix(1, count)
self.assertEqual(expected, result)
def test_count_with_pseudocounts(self):
expected = {
'T': [2, 2, 1, 2, 5, 3],
'A': [2, 3, 2, 1, 1, 3],
'C': [3, 2, 5, 3, 1, 1],
'G': [2, 2, 1, 3, 2, 2]
}
count = get_count(self.motifs)
result = add_scalar_to_matrix(1, count)
self.assertEqual(expected, result)
def test_consensus(self):
motifs = [
'AACGTA',
'CCCGTT',
'CACCTT',
'GGATTA',
'TTCCGG',
]
expected = 'CACCTA'
count_matrix = get_count(motifs)
cons_string = consensus(count_matrix)
self.assertEqual(expected, cons_string)
def profile_to_motifs(profile, dna):
"""
:param profile: profile matrix
:param dna: DNA strings
:return: best motifs
"""
k = len(list(profile.values())[0])
t = len(dna)
n = len(dna[0])
best_motifs = []
best_score = 0
for i in range(n - k + 1):
motifs = [dna[0][i:i + k]]
for j in range(1, t):
motifs.append(profile_most_probable_pattern(dna[j], k, profile))
if not best_motifs or score(motifs, consensus(get_count(motifs))) < best_score:
best_motifs = motifs
best_score = score(best_motifs, consensus(get_count(best_motifs)))
return best_motifs
def test_score(self):
motifs = [
'AACGTA',
'CCCGTT',
'CACCTT',
'GGATTA',
'TTCCGG',
]
expected = 14
count = get_count(motifs)
cons = consensus(count)
sc = score(motifs, cons)
self.assertEqual(expected, sc)
def test_count(self):
motifs = [
'AACGTA',
'CCCGTT',
'CACCTT',
'GGATTA',
'TTCCGG',
]
expected = {
'A': [1, 2, 1, 0, 0, 2],
'C': [2, 1, 4, 2, 0, 0],
'G': [1, 1, 0, 2, 1, 1],
'T': [1, 1, 0, 1, 4, 2]
}
count_matrix = get_count(motifs)
self.assertEqual(expected, count_matrix)
def test_profile(self):
motifs = [
'AACGTA',
'CCCGTT',
'CACCTT',
'GGATTA',
'TTCCGG',
]
expected = {
'A': [0.2, 0.4, 0.2, 0.0, 0.0, 0.4],
'C': [0.4, 0.2, 0.8, 0.4, 0.0, 0.0],
'G': [0.2, 0.2, 0.0, 0.4, 0.2, 0.2],
'T': [0.2, 0.2, 0.0, 0.2, 0.8, 0.4]
}
profile = get_profile(get_count(motifs))
self.assertEqual(profile, expected)
def test_greedy_motif_search_with_pseudocounts_long(self):
dna = [
'GCGCCCCGCCCGGACAGCCATGCGCTAACCCTGGCTTCGATGGCGCCGGCTCAGTTAGGGCCGGAAGTCCCCAATGTGGCAGACCTTTCGCCCCTGGCGGACGAATGACCCCAGTGGCCGGGACTTCAGGCCCTATCGGAGGGCTCCGGCGCGGTGGTCGGATTTGTCTGTGGAGGTTACACCCCAATCGCAAGGATGCATTATGACCAGCGAGCTGAGCCTGGTCGCCACTGGAAAGGGGAGCAACATC',
'CCGATCGGCATCACTATCGGTCCTGCGGCCGCCCATAGCGCTATATCCGGCTGGTGAAATCAATTGACAACCTTCGACTTTGAGGTGGCCTACGGCGAGGACAAGCCAGGCAAGCCAGCTGCCTCAACGCGCGCCAGTACGGGTCCATCGACCCGCGGCCCACGGGTCAAACGACCCTAGTGTTCGCTACGACGTGGTCGTACCTTCGGCAGCAGATCAGCAATAGCACCCCGACTCGAGGAGGATCCCG',
'ACCGTCGATGTGCCCGGTCGCGCCGCGTCCACCTCGGTCATCGACCCCACGATGAGGACGCCATCGGCCGCGACCAAGCCCCGTGAAACTCTGACGGCGTGCTGGCCGGGCTGCGGCACCTGATCACCTTAGGGCACTTGGGCCACCACAACGGGCCGCCGGTCTCGACAGTGGCCACCACCACACAGGTGACTTCCGGCGGGACGTAAGTCCCTAACGCGTCGTTCCGCACGCGGTTAGCTTTGCTGCC',
'GGGTCAGGTATATTTATCGCACACTTGGGCACATGACACACAAGCGCCAGAATCCCGGACCGAACCGAGCACCGTGGGTGGGCAGCCTCCATACAGCGATGACCTGATCGATCATCGGCCAGGGCGCCGGGCTTCCAACCGTGGCCGTCTCAGTACCCAGCCTCATTGACCCTTCGACGCATCCACTGCGCGTAAGTCGGCTCAACCCTTTCAAACCGCTGGATTACCGACCGCAGAAAGGGGGCAGGAC',
'GTAGGTCAAACCGGGTGTACATACCCGCTCAATCGCCCAGCACTTCGGGCAGATCACCGGGTTTCCCCGGTATCACCAATACTGCCACCAAACACAGCAGGCGGGAAGGGGCGAAAGTCCCTTATCCGACAATAAAACTTCGCTTGTTCGACGCCCGGTTCACCCGATATGCACGGCGCCCAGCCATTCGTGACCGACGTCCCCAGCCCCAAGGCCGAACGACCCTAGGAGCCACGAGCAATTCACAGCG',
'CCGCTGGCGACGCTGTTCGCCGGCAGCGTGCGTGACGACTTCGAGCTGCCCGACTACACCTGGTGACCACCGCCGACGGGCACCTCTCCGCCAGGTAGGCACGGTTTGTCGCCGGCAATGTGACCTTTGGGCGCGGTCTTGAGGACCTTCGGCCCCACCCACGAGGCCGCCGCCGGCCGATCGTATGACGTGCAATGTACGCCATAGGGTGCGTGTTACGGCGATTACCTGAAGGCGGCGGTGGTCCGGA',
'GGCCAACTGCACCGCGCTCTTGATGACATCGGTGGTCACCATGGTGTCCGGCATGATCAACCTCCGCTGTTCGATATCACCCCGATCTTTCTGAACGGCGGTTGGCAGACAACAGGGTCAATGGTCCCCAAGTGGATCACCGACGGGCGCGGACAAATGGCCCGCGCTTCGGGGACTTCTGTCCCTAGCCCTGGCCACGATGGGCTGGTCGGATCAAAGGCATCCGTTTCCATCGATTAGGAGGCATCAA',
'GTACATGTCCAGAGCGAGCCTCAGCTTCTGCGCAGCGACGGAAACTGCCACACTCAAAGCCTACTGGGCGCACGTGTGGCAACGAGTCGATCCACACGAAATGCCGCCGTTGGGCCGCGGACTAGCCGAATTTTCCGGGTGGTGACACAGCCCACATTTGGCATGGGACTTTCGGCCCTGTCCGCGTCCGTGTCGGCCAGACAAGCTTTGGGCATTGGCCACAATCGGGCCACAATCGAAAGCCGAGCAG',
'GGCAGCTGTCGGCAACTGTAAGCCATTTCTGGGACTTTGCTGTGAAAAGCTGGGCGATGGTTGTGGACCTGGACGAGCCACCCGTGCGATAGGTGAGATTCATTCTCGCCCTGACGGGTTGCGTCTGTCATCGGTCGATAAGGACTAACGGCCCTCAGGTGGGGACCAACGCCCCTGGGAGATAGCGGTCCCCGCCAGTAACGTACCGCTGAACCGACGGGATGTATCCGCCCCAGCGAAGGAGACGGCG',
'TCAGCACCATGACCGCCTGGCCACCAATCGCCCGTAACAAGCGGGACGTCCGCGACGACGCGTGCGCTAGCGCCGTGGCGGTGACAACGACCAGATATGGTCCGAGCACGCGGGCGAACCTCGTGTTCTGGCCTCGGCCAGTTGTGTAGAGCTCATCGCTGTCATCGAGCGATATCCGACCACTGATCCAAGTCGGGGGCTCTGGGGACCGAAGTCCCCGGGCTCGGAGCTATCGGACCTCACGATCACC',
]
k = 15
t = len(dna)
expected_score = 35
result = greedy_motif_search_with_pseudocounts(dna, k, t)
print(result)
self.assertEqual(expected_score,
score(result, consensus(get_count(result))))
def test_greedy_motif_search_0(self):
dna = [
'GGCGTTCAGGCA',
'AAGAATCAGTCA',
'CAAGGAGTTCGC',
'CACGTCAATCAC',
'CAATAATATTCG',
]
k = 3
t = len(dna)
expected = [
'CAG',
'CAG',
'CAA',
'CAA',
'CAA',
]
result = greedy_motif_search(dna, k, t)
self.assertEqual(expected, result)
cons = consensus(get_count(expected))
self.assertEqual(2, score(result, cons))
def test_greedy_motif_search_with_pseudocounts_0(self):
dna = [
'GGCGTTCAGGCA',
'AAGAATCAGTCA',
'CAAGGAGTTCGC',
'CACGTCAATCAC',
'CAATAATATTCG',
]
k = 3
t = len(dna)
expected = [
"TTC",
"ATC",
"TTC",
"ATC",
"TTC",
]
expected_score = 2
result = greedy_motif_search_with_pseudocounts(dna, k, t)
self.assertEqual(expected, result)
self.assertEqual(expected_score,
score(result, consensus(get_count(result))))
