def test_reads_below_k31_makes_empty_vcf(self):
nonvars = [Variant("Chr1", (20, 20), self.ref[0].seq[20])]
reads: Reads = simulate_reads(self.ref,
nonvars,
read_len=15,
fold_cov=1)
reads.write(self.paths.reads_out)
with mock.patch("cortex.calls._make_empty_vcf") as mock_empty_vcf:
cortex_run(
self.paths.ref_out,
[self.paths.reads_out],
self.paths.out_vcf,
mem_height=2,
tmp_directory=self.paths._tmp_dir,
sample_name="mysample",
)
mock_empty_vcf.assert_called_once_with(self.paths.out_vcf,
"mysample")
# Check the vcf gets actually made
self.setUp()
reads.write(self.paths.reads_out)
cortex_run(
self.paths.ref_out,
[self.paths.reads_out],
self.paths.out_vcf,
mem_height=2,
tmp_directory=self.paths._tmp_dir,
cleanup=False,
)
self.assertTrue(self.paths.out_vcf.exists())
def test_simulate_reads_with_variant(self):
variants = [Variant("r1", (9, 10), "AAT")]
with mock.patch("random.randint") as randint:
randint.return_value = 8
reads = simulate_reads(self.refs, variants, 7, fold_cov=3)
self.assertTrue(all([read.seq == "CAATGGA" for read in reads]))
def test_simulate_reads_no_variant(self):
nonvars = [Variant("r1", (9, 9), self.refs[0].seq[9])]
with mock.patch("random.randint") as randint:
randint.return_value = 8
reads = simulate_reads(self.refs, nonvars, 7, fold_cov=3)
expected_seq = str(self.refs[0].seq[8:8 + 7])
self.assertTrue(all([read.seq == expected_seq for read in reads]))
def test_not_enough_memory_fails(self):
refs = simulate_refs(["id1"], [10000])
reads = simulate_reads(refs, variants=list(), read_len=10)
with tmpInputFiles() as paths:
refs.write(paths.ref_out)
reads.write(paths.reads_out)
with self.assertRaises(RuntimeError):
cortex_run(
paths.ref_out,
[paths.reads_out],
paths.out_vcf,
tmp_directory=paths._tmp_dir,
mem_height=1,
)
def test_reads_with_no_var_makes_empty_vcf(self):
nonvars = [Variant("Chr1", (50, 50), self.ref[0].seq[50])]
reads: Reads = simulate_reads(self.ref,
nonvars,
read_len=40,
fold_cov=30)
reads.write(self.paths.reads_out)
with mock.patch("cortex.calls._make_empty_vcf") as mock_empty_vcf:
cortex_run(
self.paths.ref_out,
[self.paths.reads_out],
self.paths.out_vcf,
mem_height=2,
tmp_directory=self.paths._tmp_dir,
)
mock_empty_vcf.assert_called_once()
def test_reads_with_one_snp_non_empty_vcf(self):
snp_pos = 50
ref_base = self.ref[0].seq[snp_pos]
non_ref_choices = set(dna_choices).difference(set(ref_base))
assert len(non_ref_choices) == len(dna_choices) - 1
alt_base = random.choice(list(non_ref_choices))
vars = [Variant("Chr1", (snp_pos, snp_pos), alt_base)]
reads: Reads = simulate_reads(self.ref, vars, read_len=40, fold_cov=30)
reads.write(self.paths.reads_out)
cortex_run(
self.paths.ref_out,
[self.paths.reads_out],
self.paths.out_vcf,
mem_height=2,
tmp_directory=self.paths.reads_out.parent,
cleanup=False,
)
with self.paths.out_vcf.open() as vcf_out:
all_lines = vcf_out.readlines()
all_lines = [line for line in all_lines if line[0] != "#"]
self.assertTrue(len(all_lines) > 0)
def test_read_on_ref_end_passes(self):
variants = [
Variant("r1", (3, 3), "T"),
Variant("r1", (1, 1), "T"),
]
simulate_reads(self.refs, variants, 1)
def test_read_len_beyond_ref_end_fails(self):
variants = [Variant("r1", (3, 3), "T")]
with self.assertRaises(ValueError):
simulate_reads(self.refs, variants, 2)