def default_seqs():
"""Default sequence input used in all tests requiring sequence data.
Exceptions include tests in the TestSeqIO class."""
seqs = read_seqs(data_dir, filename_end='trimmed.fq', cutoff=5.8)
return seqs
def test_data_random_sampling(self):
seqs = read_seqs(data_dir, filename_end='trimmed.fq', p_discard=0.5)
for condition in seqs:
# With 1000 seqs, chance of having less than 400 or more than 600
# seqs by chance is low (at least six standard devs from mean).
assert len(seqs[condition]) > 400
assert len(seqs[condition]) < 600
def test_exact_length(self):
seqs_len10 = read_seqs(data_dir,
filename_end='trimmed.fq',
seq_len_req=10)
for condition in seqs_len10:
for seq in seqs_len10[condition]:
assert len(seq) == 10
def test_get_seqs_for_one_specific_condition(self):
seqs = read_seqs(data_dir,
filename_end='trimmed.fq',
cond_text='C0_rep1')
cond_name = 'test_C0_rep1_reads'
assert len(seqs) == 1
assert cond_name in seqs
def test_get_seqs_for_multiple_specific_conditions(self):
seqs = read_seqs(data_dir,
filename_end='trimmed.fq',
cond_text=['C0_rep1', 'C1_rep1'])
cond1_name = 'test_C0_rep1_reads'
cond2_name = 'test_C1_rep1_reads'
assert len(seqs) == 2
assert cond1_name in seqs
assert cond2_name in seqs
def test_degen(self):
seqs_degen = read_seqs(data_dir, filename_end='trimmed.fq', degen=5)
first_seqs = [
'CAC', 'AGAATGAG', 'AAAATAGCTGGAGGATC', 'ACGGGGGATGCAGAGGGGTTGTCC',
'AGTGAAAGGATAGGAAGGTCA', 'GAGTCGAGACGAGAAGGATA'
]
for i, condition in enumerate(seqs_degen):
first_seq = list(first_seqs[i])
assert seqs_degen[condition][0] == first_seq
def test_cutoff_int(self):
seqs_cutoff_int = read_seqs(data_dir,
filename_end='trimmed.fq',
cutoff=5.0)
first_seqs = [
'CTA', 'GTAGGAGA', 'GAGGTAAAATAGCTGGA', 'AAGTGACGGGGGATGCAGAGGGGT',
'GATCGAGTGAAAGGATAGGAA', 'CGGGTGAGTCGAGACGAGAA'
]
for i, condition in enumerate(seqs_cutoff_int):
first_seq = list(first_seqs[i])
assert seqs_cutoff_int[condition][0] == first_seq
def test_degen_cutoff(self):
seqs_degen_cutoff = read_seqs(data_dir,
filename_end='trimmed.fq',
degen=5,
cutoff=5.8)
first_seqs = [
('G', 'GGAAACGGGAAAGCTAAATCAAGAGA',
'GGAAACGGGAAAGCTAAATCAAGAG'), # first seq too short
('AGA', 'AG'),
('AAAATAGCTGGA', 'AAAATAGCTGG'),
('ACGGGGGATGCAGAGGGGT', 'ACGGGGGATGCAGAGGGG'),
('AGTGAAAGGATAGGAA', 'AGTGAAAGGATAGGA'),
('GAGTCGAGACGAGAA', 'GAGTCGAGACGAGA')
]
for i, condition in enumerate(seqs_degen_cutoff):
first_seq_possibilities = (list(seq) for seq in first_seqs[i])
assert seqs_degen_cutoff[condition][0] in first_seq_possibilities
def test_p_discard_greater_than_1_raises_ValueError(self):
with pytest.raises(ValueError):
read_seqs(data_dir, filename_end='trimmed.fq', p_discard=1.5)
def test_neg_p_discard_raises_ValueError(self):
with pytest.raises(ValueError):
read_seqs(data_dir, filename_end='trimmed.fq', p_discard=-0.5)
def test_neg_cutoff_raises_ValueError(self):
with pytest.raises(ValueError):
read_seqs(data_dir, filename_end='trimmed.fq', cutoff=-1)
def seqs(self):
return read_seqs(data_dir, filename_end='trimmed.fq')