def test_epitope_unequal(self):
seq1_example = 'aaAA'
seq2_example = 'aaaA'
epitope1 = Epitope(SeqRecord(Seq(seq1_example)))
epitope2 = Epitope(SeqRecord(Seq(seq2_example)))
self.assertNotEqual(epitope1, epitope2)
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 test_add_verified_region(self):
seq_example = 'AAaa'
verified_region_to_add_example = (3, 3)
expected_verified_regions = [(0, 1), (3, 3)]
expected_record_seq = 'AAaA'
epitope = Epitope(SeqRecord(Seq(seq_example)))
epitope.add_verified_region(verified_region_to_add_example)
actual_verified_regions = epitope.verified_regions
actual_record_seq = str(epitope.record.seq)
self.assertEqual(expected_verified_regions, actual_verified_regions)
self.assertEqual(expected_record_seq, actual_record_seq)
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 __parse_clstr_file(clstr_file_path: str,
records_fasta_path: str) -> List[List[Epitope]]:
with open(records_fasta_path) as records_fasta_file:
records_dict = SeqIO.to_dict(
SeqIO.parse(records_fasta_file, 'fasta'))
epitopes_clusters_lst = []
with open(clstr_file_path) as epitopes_ids_clusters_file:
curr_cluster = []
for line in epitopes_ids_clusters_file.readlines():
line = line.strip()
if line != '':
# when new cluster found appending the current cluster set and creating new one
# if the cluster set is not empty (should occur on first line)
if line.startswith(CLUSTER_PREFIX):
if len(curr_cluster) > 0:
epitopes_clusters_lst.append(curr_cluster)
curr_cluster = []
else:
epitope_id = line.split(EPITOPE_ID_PREFIX)[1].split(
EPITOPE_ID_SUFFIX)[0]
seq_record = records_dict[epitope_id]
epitope = Epitope(seq_record)
curr_cluster.append(epitope)
# adding last cluster ser
if len(curr_cluster) > 0:
epitopes_clusters_lst.append(curr_cluster)
return epitopes_clusters_lst
def test_epitope_init_verified_region3(self):
seq_example = 'AAaa'
expected_verified_regions = [(0, 1)]
epitope = Epitope(SeqRecord(Seq(seq_example)))
actual_verified_regions = epitope.verified_regions
self.assertEqual(expected_verified_regions, actual_verified_regions)
def test_init1(self):
seq_example = 'aaaA'
expected_seq = seq_example
expected_verified_regions = [(3, 3)]
actual_epitope = Epitope(SeqRecord(Seq(seq_example)))
actual_verified_regions = actual_epitope.verified_regions
self.assertEqual(expected_seq, str(actual_epitope))
self.assertEqual(expected_verified_regions, actual_verified_regions)
def __parse_records_batches_fasta_files(
records_batches_fasta_paths: List[str]) -> List[Epitope]:
raw_records = []
for records_batch_fasta_path in records_batches_fasta_paths:
with open(records_batch_fasta_path) as records_batch_file:
records_batch = [
Epitope(seq_record)
for seq_record in SeqIO.parse(records_batch_file, 'fasta')
]
raw_records.extend(records_batch)
return raw_records
def test_unequal(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('a'))),
Epitope(SeqRecord(Seq('B'))),
Epitope(SeqRecord(Seq('aa'))),
Epitope(SeqRecord(Seq('aa'))),
]
)
self.assertNotEqual(epitopes_dataset1, epitopes_dataset2)
def test_init(self):
expected_epitopes_clusters_lst = [[
Epitope(SeqRecord('AaAA')),
Epitope(SeqRecord('aaaa'))
], [Epitope(SeqRecord('bBBBB'))],
[
Epitope(SeqRecord('ddDDD')),
Epitope(SeqRecord('DDdDdD')),
Epitope(SeqRecord('DDdddD'))
]]
actual_epitopes_clusters = EpitopesClusters(EPITOPES_CLUSTERS_PATH,
EPITOPES_FASTA_PATH)
self.assertEqual(expected_epitopes_clusters_lst,
list(actual_epitopes_clusters))
def _add_verified_regions_lst(epitope: Epitope, verified_regions_lst: List[Tuple[int, int]]) -> Epitope:
for verified_region in verified_regions_lst:
epitope.add_verified_region(verified_region)
return epitope
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_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)