def setUp(self):
self.dir = tempfile.mkdtemp() + '/'
self.db = self.dir + 'TEMPDB'
self.metadata = self.dir + 'TEMPmetadata'
self.fasta = self.dir + 'TEMP'
file_header = 'Fasta,Class,Dataset\n'
file_contents = 'TEMP,A,Train\n'
with open(self.metadata, 'w') as f:
f.write('%s%s' % (file_header, file_contents))
with open(self.fasta, 'w') as f:
out = '>label1\nAAAA\n>label2\nCCCC\n>label3\nATAT\n>label4\nACGT'
f.write(out)
metadata_kwargs = {
'metadata': self.metadata,
'prefix': self.dir,
'validate': False
}
kmer_kwargs = {'k': 4, 'limit': 0}
args = {
'metadata_kwargs': metadata_kwargs,
'kmer_kwargs': kmer_kwargs,
'database': self.db,
'recount': True,
'validate': False,
'complete_count': False
}
self.data = get_kmer(**args)
self.correct_features = np.array(['AAAA', 'ACGT', 'ATAT', 'CCCC'])
def test_y_test_complete(self):
self.args['complete_count'] = True
self.args['database'] = self.db_complete
self.args['kmer_kwargs'] = {'k': 2}
data, features, files, le = get_kmer(**self.args)
val = False
if np.array_equal(le.inverse_transform(data[3]), self.correct_y_test):
val = True
self.assertTrue(val)
def test_y_train(self):
data, features, files, le = get_kmer(**self.args)
count1 = 0
count2 = 0
for elem in self.correct_y_train:
count2 += 1
if elem in le.inverse_transform(data[1]):
count1 += 1
for elem in le.inverse_transform(data[1]):
count2 += 1
if elem in self.correct_y_train:
count1 += 1
self.assertEqual(count1, count2)
def test_x_train(self):
data, features, files, le = get_kmer(**self.args)
count1 = 0
count2 = 0
for elem in self.correct_x_train:
count2 += 1
if elem in data[0]:
count1 += 1
for elem in data[0]:
count2 += 1
if elem in self.correct_x_train:
count1 += 1
self.assertEqual(count1, count2)
def test_x_train_complete(self):
self.args['complete_count'] = True
self.args['database'] = self.db_complete
self.args['kmer_kwargs'] = {'k': 2}
data, features, files, le = get_kmer(**self.args)
count1 = 0
count2 = 0
for elem in self.correct_x_train:
count2 += 1
if elem in data[0]:
count1 += 1
for elem in data[0]:
count2 += 1
if elem in self.correct_x_train:
count1 += 1
self.assertEqual(count1, count2)
def test_y_test(self):
data, features, files, le = get_kmer(**self.args)
val = False
if np.array_equal(le.inverse_transform(data[3]), self.correct_y_test):
val = True
self.assertTrue(val)
def test_x_test(self):
data, features, files, le = get_kmer(**self.args)
val = False
if np.array_equal(data[2], self.correct_x_test):
val = True
self.assertTrue(val)