def test_get_random_string(self):
"""Test get_random_string"""
with captured_output() as (_, _):
test_string = get_random_string(5, "AAAA")
self.assertEqual(test_string, "AAAAA")
test_string = get_random_string(5, "ATGC")
self.assertEqual(len(test_string), 5)
self.assertTrue(set(test_string).issubset(set("ATGC")))
def test_get_event_mean_gaussian_parameters(self):
for x in range(10):
kmer = get_random_string(5, chars="ATGC")
mu, sigma = self.model.get_event_mean_gaussian_parameters(kmer)
mu, sigma = self.model.get_event_mean_gaussian_parameters("TTTTT")
self.assertEqual(self.model.event_model["means"][-1], mu)
self.assertEqual(self.model.event_model["SDs"][-1], sigma)
def test_get_event_sd_inv_gaussian_parameters(self):
for x in range(10):
kmer = get_random_string(5, chars="ATGC")
mu, sigma = self.model.get_event_sd_inv_gaussian_parameters(kmer)
mean, lambda1 = self.model.get_event_sd_inv_gaussian_parameters(
"TTTTT")
self.assertEqual(self.model.event_model["noise_means"][-1], mean)
self.assertEqual(self.model.event_model["noise_lambdas"][-1], lambda1)
def test_count_kmers(self):
self.assertRaises(AssertionError, count_kmers, '', 1)
self.assertRaises(AssertionError, count_kmers, 'asdf', 0)
for x in range(10):
rand_len = np.random.randint(1, 1000)
random_string = get_random_string(rand_len)
rand_k = np.random.randint(1, 10)
kmer_counts = count_kmers(random_string, rand_k)
for kmer in kmer_counts.keys():
self.assertEqual(kmer_counts[kmer], len([x for x in find_substring_indices(random_string, kmer)]),
"random_string: {} \n kmer: {}".format(random_string, kmer))
def test_kmer_iterator(self):
error_test = kmer_iterator('', 1)
self.assertRaises(AssertionError, error_test.__next__)
error_test = kmer_iterator('asdf', 0)
self.assertRaises(AssertionError, error_test.__next__)
for x in range(10):
rand_len = np.random.randint(1, 1000)
random_string = get_random_string(rand_len)
rand_k = np.random.randint(1, 10)
kmer_generator = kmer_iterator(random_string, rand_k)
for i, kmer in enumerate(kmer_generator):
self.assertEqual(rand_k, len(kmer))
self.assertEqual(random_string[i:i + rand_k], kmer)
def test_reverse_complement(self):
rev_comp = ReverseComplement(find="ACGTMKRYBVDHNacgtmkrybvdhn", replace="TGCAKMYRVBHDNtgcakmyrvbhdn")
for x in range(10):
rand_len = np.random.randint(0, 1000)
random_dna = get_random_string(rand_len, chars=list(set("ACGTMKRYBVDHN")))
self.assertEqual(reverse_complement(random_dna, reverse=True, complement=True),
rev_comp.reverse_complement(random_dna))
self.assertEqual(reverse_complement(random_dna, reverse=False, complement=True),
rev_comp.complement(random_dna))
self.assertEqual(reverse_complement(random_dna, reverse=True, complement=False),
rev_comp.reverse(random_dna))
self.assertEqual(reverse_complement(random_dna, reverse=False, complement=False),
random_dna)
def test_log_event_mean_gaussian_probability_match(self):
def emissions_signal_logGaussPdf(x, mu, sigma):
log_inv_sqrt_2pi = -0.91893853320467267
l_sigma = np.log(sigma)
a = (x - mu) / sigma
# // returns Log-space
return log_inv_sqrt_2pi - l_sigma + (-0.5 * a * a)
for x in range(10):
kmer = get_random_string(5, chars="ATGC")
mu, sigma = self.model.get_event_mean_gaussian_parameters(kmer)
prob = self.model.log_event_mean_gaussian_probability_match(
50, kmer)
self.assertAlmostEqual(prob,
emissions_signal_logGaussPdf(50, mu, sigma))
def test_log_event_sd_inv_gaussian_probability_match(self):
def emissions_signal_logInvGaussPdf(eventNoise, modelNoiseMean,
modelNoiseLambda):
l_twoPi = 1.8378770664093453 # // log(2*pi)
l_eventNoise = np.log(eventNoise)
a = (eventNoise - modelNoiseMean) / modelNoiseMean
l_modelNoseLambda = np.log(modelNoiseLambda)
# // returns Log-space
return (l_modelNoseLambda - l_twoPi - 3 * l_eventNoise -
(modelNoiseLambda * a * a / eventNoise)) / 2
for x in range(10):
kmer = get_random_string(5, chars="ATGC")
mu, lambda1 = self.model.get_event_sd_inv_gaussian_parameters(kmer)
prob = self.model.log_event_sd_inv_gaussian_probability_match(
2, kmer)
self.assertAlmostEqual(
prob, emissions_signal_logInvGaussPdf(2, mu, lambda1))
def test_get_kmer_index(self):
all_kmers = [x for x in all_string_permutations("ATGC", 5)]
for x in range(10):
kmer = get_random_string(5, chars="ATGC")
self.assertEqual(all_kmers.index(kmer),
self.model.get_kmer_index(kmer))
