def test_find_longest_orf_stopsites(self):
# check that the stop site is calculated as the * for seqs with it,
# and the last AA for those without
stop_loc_with_stop = find_longest_orfs(['AMEATBALL*'])[2]
stop_loc_without_stop = find_longest_orfs(['AMEATBALL'])[2]
self.assertEqual(stop_loc_with_stop, stop_loc_without_stop + 1)
def test_convert_nt_output_format(self):
# tests that a length 3 tupple output, and each is the correct numpy
# array type
sequences = read_fasta('test_data/test_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=False)
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_frames)
# filter data by minimum orf length
keep = orf_length >= 6
aa_frames, frame, strand, seq_length_nt, ids, seq_length, start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length = filter_objects(
keep, aa_frames, frame, strand, seq_length_nt, ids, seq_length,
start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length)
output = convert_start_stop_to_nt(start_sites, stop_sites,
seq_length_nt, orf_length, frame,
last_aa_is_stop)
t_len = len(output) == 3
# test numpy types of all outputs
t0 = output[0].dtype == 'int64'
t1 = output[1].dtype == 'int64'
t2 = output[2].dtype == 'int64'
all_right_types = t0 and t1 and t2 and t_len
self.assertTrue(all_right_types)
def test_add_upstream_aa_exactupstream(self):
# test with upstream AAs of exactly min_upstream_length
aa_sequence = np.array(['BEARMEATBALLS'])
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_sequence)
output = add_upstream_aas(aa_sequence,
stop_sites,
start_sites,
orf_sequence,
orf_length,
min_upstream_length=4)
self.assertEqual(output[0], 'BEARMEATBALLS')
def test_add_upstream_aa_noupstream(self):
# test with no viable upstream AAs
aa_sequence = np.array(['BEAREATS*MEATBALLS'])
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_sequence)
output = add_upstream_aas(aa_sequence,
stop_sites,
start_sites,
orf_sequence,
orf_length,
min_upstream_length=5)
self.assertEqual(output[0], 'MEATBALLS')
def test_add_upstream_aa(self):
# test expected output
aa_sequence = np.array(['ALONGERUPSTREAMMEATBALL'])
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_sequence)
output = add_upstream_aas(aa_sequence,
stop_sites,
start_sites,
orf_sequence,
orf_length,
min_upstream_length=5)
self.assertEqual(output[0], 'ALONGERUPSTREAMMEATBALL')
def test_find_longest_orf_output_format(self):
# tests that a length 5 tupple output, and each is the correct numpy
# array type
long_orf = find_longest_orfs(['AMEATBALL'])
t_len = len(long_orf) == 5
# test numpy types of all outputs
t0 = long_orf[0].dtype == '<U8'
t1 = long_orf[1].dtype == 'int64'
t2 = long_orf[2].dtype == 'int64'
t3 = long_orf[3].dtype == 'int64'
t4 = long_orf[4].dtype == 'bool'
all_right_types = t0 and t1 and t2 and t3 and t4 and t_len
self.assertTrue(all_right_types)
def test_convert_utr_nt(self):
sequences = read_fasta('test_data/test_frames.fa')
ids, aa_frames, frame, strand, seq_length_nt, seq_length = translate_all_frames(
sequences, both_strands=False)
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_frames)
# filter data by minimum orf length
keep = orf_length >= 6
aa_frames, frame, strand, seq_length_nt, ids, seq_length, start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length = filter_objects(
keep, aa_frames, frame, strand, seq_length_nt, ids, seq_length,
start_sites, stop_sites, orf_sequence, last_aa_is_stop, orf_length)
start_site_nt, stop_site_nt, utr3_length = convert_start_stop_to_nt(
start_sites, stop_sites, seq_length_nt, orf_length, frame,
last_aa_is_stop)
self.assertTrue(np.all(utr3_length == np.array([5, 4, 3])))
def test_add_upstream_aa_multi(self):
# test with multiple inputs
aa_sequence = np.array([
'ALONGERUPSTREAMMEATBALL', 'TWODOZENMEATBALLS',
'BROWNBEARMAULSGIANTSQUID'
])
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_sequence)
output = add_upstream_aas(aa_sequence,
stop_sites,
start_sites,
orf_sequence,
orf_length,
min_upstream_length=5)
self.assertTrue(
np.all(output[0] == np.array([
'ALONGERUPSTREAMMEATBALL', 'TWODOZENMEATBALLS',
'BROWNBEARMAULSGIANTSQUID'
])))
def test_add_upstream_aa_output(self):
# check all outputs generated and all in correct type
aa_sequence = np.array(['ALONGERUPSTREAMMEATBALL'])
orf_sequence, start_sites, stop_sites, orf_length, last_aa_is_stop = find_longest_orfs(
aa_sequence)
output = add_upstream_aas(aa_sequence,
stop_sites,
start_sites,
orf_sequence,
orf_length,
min_upstream_length=5)
t_len = len(output) == 3
# test numpy types of all outputs
t0 = output[0].dtype.type == np.str_
t1 = output[1].dtype == 'int64'
t2 = output[2].dtype == 'int64'
all_right_types = t0 and t1 and t2 and t_len
self.assertTrue(all_right_types)
def test_find_longest_orf_multiple(self):
input = ['AMEATBALL*', 'TWOMEATBALLS']
result = find_longest_orfs(input)
self.assertEqual(len(result[0]), len(input))
def test_find_longest_orf_trimmed(self):
# check that the last * is trimmed from the orf sequence
self.assertEqual(find_longest_orfs(['AMEATBALL*'])[0], ['MEATBALL'])