def test_save_snps_vcf_false_positive_build(self):
with tempfile.TemporaryDirectory() as tmpdir1:
snps = SNPs("tests/input/testvcf.vcf", output_dir=tmpdir1)
r = Resources()
r._reference_sequences["GRCh37"] = {}
output = os.path.join(tmpdir1, "vcf_GRCh37.vcf")
with tempfile.TemporaryDirectory() as tmpdir2:
dest = os.path.join(tmpdir2, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
r._reference_sequences["GRCh37"]["1"] = seq
self.assertEqual(snps.save(vcf=True), output)
s = ""
with open(output, "r") as f:
for line in f.readlines():
if "snps v" in line:
s += '##source="vcf; snps v1.2.3.post85.dev0+gb386302; https://pypi.org/project/snps/"\n'
else:
s += line
with open(output, "w") as f:
f.write(s)
self.run_parsing_tests_vcf(output)
def test_reference_sequence_generic_load_sequence(self):
with tempfile.TemporaryDirectory() as tmpdir:
dest = os.path.join(tmpdir, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
self.assertEqual(seq.ID, "1")
self.assertEqual(seq.chrom, "1")
self.assertEqual(seq.path, dest)
np.testing.assert_array_equal(
seq.sequence,
np.array(
bytearray(
"NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNAGGCCGGACNNNNNNNN",
encoding="utf-8",
errors="strict",
),
dtype=np.uint8,
),
)
self.assertListEqual(list("AGGCCGGAC"),
list(map(chr, seq.sequence[100:109])))
self.assertEqual(seq.md5, "6ac6176535ad0e38aba2d05d786c39b6")
self.assertEqual(seq.start, 1)
self.assertEqual(seq.end, 117)
self.assertEqual(seq.length, 117)
def test_save_snps_vcf_discrepant_pos(self):
s = SNPs("tests/input/testvcf.vcf")
r = Resources()
r._reference_sequences["GRCh37"] = {}
with tempfile.TemporaryDirectory() as tmpdir:
dest = os.path.join(tmpdir, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
r._reference_sequences["GRCh37"]["1"] = seq
# create discrepant SNPs by setting positions outside reference sequence
s._snps.loc["rs1", "pos"] = 0
s._snps.loc["rs17", "pos"] = 118
self.assertEqual(os.path.relpath(s.save(vcf=True)),
f"output{os.sep}vcf_GRCh37.vcf")
pd.testing.assert_frame_equal(
s.discrepant_vcf_position,
self.create_snp_df(
rsid=["rs1", "rs17"],
chrom=["1", "1"],
pos=[0, 118],
genotype=["AA", np.nan],
),
check_exact=True,
)
expected = self.generic_snps_vcf().drop(["rs1", "rs17"])
self.run_parsing_tests_vcf("output/vcf_GRCh37.vcf", snps_df=expected)
def test_reference_sequence_generic_load_sequence(self):
with open("tests/input/generic.fa", "rb") as f_in:
with atomic_write("tests/input/generic.fa.gz",
mode="wb",
overwrite=True) as f_out:
with gzip.open(f_out, "wb") as f_gzip:
shutil.copyfileobj(f_in, f_gzip)
seq = ReferenceSequence(ID="1", path="tests/input/generic.fa.gz")
assert seq.ID == "1"
assert seq.chrom == "1"
assert seq.path == "tests/input/generic.fa.gz"
np.testing.assert_array_equal(
seq.sequence,
np.array(
bytearray(
"NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNAGGCCGGACNNNNNNNN",
encoding="utf-8",
errors="strict",
),
dtype=np.uint8,
),
)
assert list("AGGCCGGAC") == list(map(chr, seq.sequence[100:109]))
assert seq.md5 == "6ac6176535ad0e38aba2d05d786c39b6"
assert seq.start == 1
assert seq.end == 117
assert seq.length == 117
def test_save_snps_vcf_phased(self):
# read phased data
s = SNPs("tests/input/testvcf_phased.vcf")
# setup resource to use test FASTA reference sequence
r = Resources()
r._reference_sequences["GRCh37"] = {}
with open("tests/input/generic.fa", "rb") as f_in:
with atomic_write("tests/input/generic.fa.gz",
mode="wb",
overwrite=True) as f_out:
with gzip.open(f_out, "wb") as f_gzip:
shutil.copyfileobj(f_in, f_gzip)
seq = ReferenceSequence(ID="1", path="tests/input/generic.fa.gz")
r._reference_sequences["GRCh37"]["1"] = seq
# save phased data to VCF
assert os.path.relpath(
s.save_snps(vcf=True)) == "output/vcf_GRCh37.vcf"
# read saved VCF
s = SNPs("output/vcf_GRCh37.vcf")
assert s.phased
pd.testing.assert_frame_equal(s.snps, self.generic_snps_vcf())
def test_save_snps_vcf(self):
s = SNPs("tests/input/testvcf.vcf")
r = Resources()
r._reference_sequences["GRCh37"] = {}
with tempfile.TemporaryDirectory() as tmpdir:
dest = os.path.join(tmpdir, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
r._reference_sequences["GRCh37"]["1"] = seq
self.assertEqual(os.path.relpath(s.save(vcf=True)),
f"output{os.sep}vcf_GRCh37.vcf")
self.run_parsing_tests_vcf("output/vcf_GRCh37.vcf")
def test_save_snps_vcf(self):
with tempfile.TemporaryDirectory() as tmpdir1:
s = SNPs("tests/input/testvcf.vcf", output_dir=tmpdir1)
r = Resources()
r._reference_sequences["GRCh37"] = {}
output = os.path.join(tmpdir1, "vcf_GRCh37.vcf")
with tempfile.TemporaryDirectory() as tmpdir2:
dest = os.path.join(tmpdir2, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
r._reference_sequences["GRCh37"]["1"] = seq
self.assertEqual(s.save(vcf=True), output)
self.run_parsing_tests_vcf(output)
def test_save_snps_vcf_phased(self):
# read phased data
s = SNPs("tests/input/testvcf_phased.vcf")
# setup resource to use test FASTA reference sequence
r = Resources()
r._reference_sequences["GRCh37"] = {}
with tempfile.TemporaryDirectory() as tmpdir:
dest = os.path.join(tmpdir, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
r._reference_sequences["GRCh37"]["1"] = seq
# save phased data to VCF
self.assertEqual(os.path.relpath(s.save(vcf=True)),
f"output{os.sep}vcf_GRCh37.vcf")
# read saved VCF
self.run_parsing_tests_vcf("output/vcf_GRCh37.vcf", phased=True)
def test_save_snps_vcf_discrepant_pos(self):
with tempfile.TemporaryDirectory() as tmpdir1:
s = SNPs("tests/input/testvcf.vcf", output_dir=tmpdir1)
r = Resources()
r._reference_sequences["GRCh37"] = {}
output = os.path.join(tmpdir1, "vcf_GRCh37.vcf")
with tempfile.TemporaryDirectory() as tmpdir2:
dest = os.path.join(tmpdir2, "generic.fa.gz")
gzip_file("tests/input/generic.fa", dest)
seq = ReferenceSequence(ID="1", path=dest)
r._reference_sequences["GRCh37"]["1"] = seq
# create discrepant SNPs by setting positions outside reference sequence
s._snps.loc["rs1", "pos"] = 0
s._snps.loc["rs17", "pos"] = 118
# esnure this is the right type after manual tweaking
s._snps = s._snps.astype({"pos": np.uint32})
self.assertEqual(s.save(vcf=True), output)
pd.testing.assert_frame_equal(
s.discrepant_vcf_position,
self.create_snp_df(
rsid=["rs1", "rs17"],
chrom=["1", "1"],
pos=[0, 118],
genotype=["AA", np.nan],
),
check_exact=True,
)
expected = self.generic_snps_vcf().drop(["rs1", "rs17"])
self.run_parsing_tests_vcf(output, snps_df=expected)