def test_write1(self):
expected_output_file_text = \
""">seq1
aaaA
>seq2
aaAA
>seq3
B
>seq1
aaaA
>seq4
aAAa
>seq5
bBBbb
>seq6
bbbBB
>seq7
bbBBb
>seq8
cccCC
>seq9
ccCCc
>seq10
Dd
>seq11
dD
"""
epitopes_dataset = EpitopesDataset(EPITOPES_BATCHES_PATHS)
epitopes_dataset.write(WRITE_DATASET_RES_PATH)
with open(WRITE_DATASET_RES_PATH) as output_file:
actual_output_file_text = output_file.read()
self.assertEqual(expected_output_file_text,
actual_output_file_text)
def test_count_verified_regions(self):
expected_verified_regions_count = 12
epitopes_dataset = EpitopesDataset(EPITOPES_BATCHES_PATHS)
actual_verified_regions_count = epitopes_dataset.count_verified_regions()
self.assertEqual(expected_verified_regions_count, actual_verified_regions_count)
def load_data(self, fasta_paths_lst: List[str]):
"""
Loads input epitopes batch files
Parameters
----------
fasta_paths_lst : List[str]
List of input epitopes batch files in fasta format
"""
epitopes_dataset = EpitopesDataset(fasta_paths_lst)
epitopes_dataset.merge_identical_seqs()
self.__epitopes_dataset = epitopes_dataset
def test_merge_identical_seqs(self):
expected_epitopes = \
[
add_verified_regions_lst(Epitope(SeqRecord(Seq('aaaA'))), [(2, 3), (3, 3), (1, 2)]),
Epitope(SeqRecord(Seq('B'))),
add_verified_regions_lst(Epitope(SeqRecord(Seq('bBBbb'))), [(3, 4), (2, 3)]),
add_verified_regions_lst(Epitope(SeqRecord(Seq('cccCC'))), [(2, 3)]),
add_verified_regions_lst(Epitope(SeqRecord(Seq('Dd'))), [(1, 1)])
]
epitopes_dataset = EpitopesDataset(EPITOPES_BATCHES_PATHS)
epitopes_dataset.merge_identical_seqs()
actual_epitopes = list(epitopes_dataset)
self.assertEqual(expected_epitopes, actual_epitopes)
def get_epitopes_with_max_verified_regions(
epitopes_clusters: EpitopesClusters) -> EpitopesDataset:
"""
Gets a dataset of the epitope records with maximum verified regions from each cluster
Parameters
----------
epitopes_clusters: model.EpitopesDataset.EpitopesDataset
Returns
-------
max_verified_regions_epitopes_dataset : model.EpitopesDataset.EpitopesDataset
Dataset of the epitope records with maximum verified regions from each cluster
"""
remaining_epitopes = []
for cluster in epitopes_clusters:
epitope_with_max_verified_regions = cluster[0]
max_verified_regions = len(cluster[0].verified_regions)
for epitope in cluster:
if len(epitope.verified_regions) > max_verified_regions:
epitope_with_max_verified_regions = epitope
max_verified_regions = len(epitope.verified_regions)
remaining_epitopes.append(epitope_with_max_verified_regions)
return EpitopesDataset(remaining_epitopes)
def __init__(self, docker_client: DockerClient, temp_output_dir_path: str, cd_hit_img_id: str):
self.__epitopes_dataset: EpitopesDataset = EpitopesDataset([])
self.__temp_output_dir = temp_output_dir_path
self.__docker_client = docker_client
self.__cd_hit_img_id = cd_hit_img_id
makedirs(self.__temp_output_dir, exist_ok=True)
def test_equal(self):
epitopes_dataset1 = EpitopesDataset(
[
Epitope(SeqRecord(Seq('a'))),
Epitope(SeqRecord(Seq('A'))),
Epitope(SeqRecord(Seq('aa'))),
Epitope(SeqRecord(Seq('aa'))),
]
)
epitopes_dataset2 = EpitopesDataset(
[
Epitope(SeqRecord(Seq('aa'))),
Epitope(SeqRecord(Seq('A'))),
Epitope(SeqRecord(Seq('a'))),
]
)
self.assertEqual(epitopes_dataset1, epitopes_dataset2)
def test_get_epitopes_with_max_verified_regions(self):
expected_epitopes_dataset = EpitopesDataset([
Epitope(SeqRecord(Seq('AaAA'))),
Epitope(SeqRecord(Seq('bBBBB'))),
Epitope(SeqRecord(Seq('DDdDdD')))
])
epitopes_clusters = EpitopesClusters(EPITOPES_CLUSTERS1_PATH,
EPITOPES_FASTA1_PATH)
actual_epitopes_dataset = get_epitopes_with_max_verified_regions(
epitopes_clusters)
self.assertEqual(expected_epitopes_dataset, actual_epitopes_dataset)
def split_epitopes_clusters_to_cv_datasets(
epitopes_clusters: EpitopesClusters,
cv_fold: int,
shuffle_clusters=True) -> List[EpitopesDataset]:
"""
Splitting the epitopes records to CV datasets
Parameters
----------
epitopes_clusters : model.EpitopesDataset.EpitopesDataset
cv_fold : int
shuffle_clusters : bool
Returns
-------
epitopes_cv_datasets : List[model.EpitopesDataset.EpitopesDataset]
CV epitope records datasets
"""
epitopes_cv_groups = [[]]
set_approximate_size = -(-epitopes_clusters.get_num_of_epitopes() //
cv_fold)
if shuffle_clusters:
epitopes_clusters = list(epitopes_clusters)
random.shuffle(epitopes_clusters)
curr_groups_ind = 0
for cluster in epitopes_clusters:
epitopes_cv_groups[curr_groups_ind].extend(cluster)
if len(epitopes_cv_groups[curr_groups_ind]) >= set_approximate_size:
epitopes_cv_groups.append([])
curr_groups_ind += 1
if len(epitopes_cv_groups[-1]) == 0:
epitopes_cv_groups = epitopes_cv_groups[:-1]
epitopes_cv_datasets = [
EpitopesDataset(epitopes_group)
for epitopes_group in epitopes_cv_groups
]
return epitopes_cv_datasets
def test_init_epitopes_dataset(self):
expected_epitopes = \
[
Epitope(SeqRecord(Seq('aaaA'))),
Epitope(SeqRecord(Seq('aaAA'))),
Epitope(SeqRecord(Seq('B'))),
Epitope(SeqRecord(Seq('aaaA'))),
Epitope(SeqRecord(Seq('aAAa'))),
Epitope(SeqRecord(Seq('bBBbb'))),
Epitope(SeqRecord(Seq('bbbBB'))),
Epitope(SeqRecord(Seq('bbBBb'))),
Epitope(SeqRecord(Seq('cccCC'))),
Epitope(SeqRecord(Seq('ccCCc'))),
Epitope(SeqRecord(Seq('Dd'))),
Epitope(SeqRecord(Seq('dD')))
]
actual_epitopes_dataset = EpitopesDataset(EPITOPES_BATCHES_PATHS)
self.assertEqual(expected_epitopes, list(actual_epitopes_dataset))
def test_write2(self):
expected_output_file_text = \
""">seq1
aAAA
>seq3
B
>seq5
bBBBB
>seq8
ccCCC
>seq10
DD
"""
epitopes_dataset = EpitopesDataset(EPITOPES_BATCHES_PATHS)
epitopes_dataset.merge_identical_seqs()
epitopes_dataset.write(WRITE_DATASET_RES_PATH)
with open(WRITE_DATASET_RES_PATH) as output_file:
actual_output_file_text = output_file.read()
self.assertEqual(expected_output_file_text,
actual_output_file_text)
def test_split_epitopes_clusters_to_cv_groups_cv10(self):
cv_fold = 10
expected_epitopes_cv_datasets = [
EpitopesDataset([
Epitope(SeqRecord(Seq('aaaAA'))),
Epitope(SeqRecord(Seq('Aaa'))),
Epitope(SeqRecord(Seq('bbbBC'))),
Epitope(SeqRecord(Seq('cccaaCd'))),
Epitope(SeqRecord(Seq('DcDcDc'))),
Epitope(SeqRecord(Seq('AAAAA')))
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('aaaG'))),
Epitope(SeqRecord(Seq('GCAcGcGa'))),
Epitope(SeqRecord(Seq('aCGPfpc'))),
Epitope(SeqRecord(Seq('cccccCCCccc'))),
Epitope(SeqRecord(Seq('GgG'))),
Epitope(SeqRecord(Seq('DDDDD')))
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('EEeeeGGGDDD'))),
Epitope(SeqRecord(Seq('BBBbbb'))),
Epitope(SeqRecord(Seq('NMnMnM'))),
Epitope(SeqRecord(Seq('KPkgK'))),
Epitope(SeqRecord(Seq('AAAaaA'))),
Epitope(SeqRecord(Seq('AAAaaaA')))
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('BBBbbBbB'))),
Epitope(SeqRecord(Seq('CCCccC'))),
Epitope(SeqRecord(Seq('GGGggGG'))),
Epitope(SeqRecord(Seq('CcCcCc'))),
Epitope(SeqRecord(Seq('cCcccc'))),
Epitope(SeqRecord(Seq('ccCccC'))),
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('CccCC'))),
Epitope(SeqRecord(Seq('cccCCcC'))),
Epitope(SeqRecord(Seq('CccCCCccC'))),
Epitope(SeqRecord(Seq('CcccCCcccc'))),
Epitope(SeqRecord(Seq('cccCCCccC'))),
Epitope(SeqRecord(Seq('cccCccccc')))
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('aaAAAaaAA'))),
Epitope(SeqRecord(Seq('BBBBbbBB'))),
Epitope(SeqRecord(Seq('bbbBBBBB'))),
Epitope(SeqRecord(Seq('GGGGgg'))),
Epitope(SeqRecord(Seq('GGGG')))
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('TTTtTTT'))),
Epitope(SeqRecord(Seq('HHHHHHhhh'))),
Epitope(SeqRecord(Seq('HHHhhhhKK'))),
Epitope(SeqRecord(Seq('kkkkKKKk'))),
Epitope(SeqRecord(Seq('UUUuuuU'))),
Epitope(SeqRecord(Seq('GFgGgF'))),
Epitope(SeqRecord(Seq('CCCcCBBb'))),
Epitope(SeqRecord(Seq('mmmmmmMMMmm'))),
Epitope(SeqRecord(Seq('BBbb')))
]),
EpitopesDataset([
Epitope(SeqRecord(Seq('GGGg'))),
Epitope(SeqRecord(Seq('AAaa'))),
Epitope(SeqRecord(Seq('AAAa')))
])
]
epitopes_clusters = EpitopesClusters(EPITOPES_CLUSTERS2_PATH,
EPITOPES_FASTA2_PATH)
actual_epitopes_cv_datasets = split_epitopes_clusters_to_cv_datasets(
epitopes_clusters, cv_fold, shuffle_clusters=False)
self.assertEqual(expected_epitopes_cv_datasets,
actual_epitopes_cv_datasets)