def test_save_discrepant_genotypes(self):
sc = SNPsCollection()
sc.load_snps(["tests/input/NCBI36.csv", "tests/input/GRCh37.csv"])
assert len(sc.discrepant_genotypes) == 1
discrepant_genotypes_file = sc.save_discrepant_genotypes()
assert (os.path.relpath(discrepant_genotypes_file) ==
"output/discrepant_genotypes.csv")
assert os.path.exists(discrepant_genotypes_file)
def test_save_discrepant_snps_specify_file(self):
sc = SNPsCollection()
sc.load_snps(["tests/input/NCBI36.csv", "tests/input/GRCh37.csv"])
assert len(sc.discrepant_snps) == 4
discrepant_snps_file = sc.save_discrepant_snps("discrepant_snps.csv")
assert os.path.relpath(
discrepant_snps_file) == "output/discrepant_snps.csv"
assert os.path.exists(discrepant_snps_file)
def test_load_snps_assembly_mismatch(self):
sc = SNPsCollection()
sc.load_snps(["tests/input/NCBI36.csv", "tests/input/GRCh37.csv"])
assert not os.path.exists("output/ind_discrepant_positions_1.csv")
assert not os.path.exists("output/ind_discrepant_genotypes_1.csv")
assert len(sc.discrepant_positions) == 4
assert len(sc.discrepant_genotypes) == 1
pd.testing.assert_frame_equal(sc.snps,
self.snps_NCBI36_discrepant_snps())
def test_source_lineage_file(self):
sc = SNPsCollection("tests/input/GRCh37.csv")
assert sc.source == "generic"
sc.load_snps("tests/input/23andme.txt")
assert sc.source == "generic, 23andMe"
file = sc.save_snps()
s = SNPs(file)
assert s.source == "generic, 23andMe"
pd.testing.assert_frame_equal(sc.snps, s.snps)
def test_source_lineage_file_gzip(self):
sc = SNPsCollection("tests/input/GRCh37.csv")
assert sc.source == "generic"
sc.load_snps("tests/input/23andme.txt")
assert sc.source == "generic, 23andMe"
file = sc.save_snps()
with open(file, "rb") as f_in:
with atomic_write(file + ".gz", mode="wb",
overwrite=True) as f_out:
with gzip.open(f_out, "wb") as f_gzip:
shutil.copyfileobj(f_in, f_gzip)
s = SNPs(file + ".gz")
assert s.source == "generic, 23andMe"
pd.testing.assert_frame_equal(sc.snps, s.snps)
def test___repr__snps_collection(self):
sc = SNPsCollection()
assert "SNPsCollection(name='')" == sc.__repr__()
def test_save_discrepant_snps_exception(self):
sc = SNPsCollection()
sc._discrepant_snps = "invalid"
assert not sc.save_discrepant_snps()
def test_save_discrepant_snps_no_discrepant_snps(self):
sc = SNPsCollection()
assert len(sc.discrepant_snps) == 0
assert not sc.save_discrepant_snps()
def test_merging_files_discrepant_snps(self):
df = pd.read_csv(
"tests/input/discrepant_snps.csv",
skiprows=1,
na_values="--",
names=[
"rsid",
"chrom",
"pos_file1",
"pos_file2",
"genotype_file1",
"genotype_file2",
"discrepant_position",
"discrepant_genotype",
"expected_position",
"expected_genotype",
],
index_col=0,
dtype={
"chrom": object,
"pos_file1": np.int64,
"pos_file2": np.int64,
"discrepant_position": bool,
"discrepant_genotype": bool,
},
)
df1 = df[["chrom", "pos_file1", "genotype_file1"]]
df2 = df[["chrom", "pos_file2", "genotype_file2"]]
df1.to_csv(
"tests/input/discrepant_snps1.csv",
na_rep="--",
header=["chromosome", "position", "genotype"],
)
df2.to_csv(
"tests/input/discrepant_snps2.csv",
na_rep="--",
header=["chromosome", "position", "genotype"],
)
sc = SNPsCollection([
"tests/input/discrepant_snps1.csv",
"tests/input/discrepant_snps2.csv"
])
expected = df[[
"chrom",
"discrepant_position",
"discrepant_genotype",
"expected_position",
"expected_genotype",
]]
expected = expected.rename(columns={
"expected_position": "pos",
"expected_genotype": "genotype"
})
expected_snps = SNPs()
expected_snps._snps = expected
expected_snps.sort_snps()
expected = expected_snps.snps
pd.testing.assert_index_equal(
sc.discrepant_positions.index,
expected.loc[expected["discrepant_position"] == True].index,
)
pd.testing.assert_index_equal(
sc.discrepant_genotypes.index,
expected.loc[expected["discrepant_genotype"] == True].index,
)
pd.testing.assert_series_equal(sc.snps["pos"], expected["pos"])
pd.testing.assert_series_equal(sc.snps["genotype"],
expected["genotype"])
def test_load_snps_invalid_file(self):
sc = SNPsCollection()
with atomic_write("tests/input/empty.txt", mode="w", overwrite=True):
pass
sc.load_snps(["tests/input/GRCh37.csv", "tests/input/empty.txt"])
pd.testing.assert_frame_equal(sc.snps, self.snps_GRCh37())
def test_load_snps_non_existent_file(self):
sc = SNPsCollection()
sc.load_snps(
["tests/input/GRCh37.csv", "tests/input/non_existent_file.csv"])
pd.testing.assert_frame_equal(sc.snps, self.snps_GRCh37())
def test_discrepant_snps(self):
sc = SNPsCollection()
sc.load_snps(["tests/input/NCBI36.csv", "tests/input/GRCh37.csv"])
assert len(sc.discrepant_snps) == 4
def test_load_snps_None(self):
sc = SNPsCollection()
with self.assertRaises(TypeError):
sc.load_snps(None)
def test_load_snps_list(self):
sc = SNPsCollection()
sc.load_snps(["tests/input/GRCh37.csv", "tests/input/GRCh37.csv"])
pd.testing.assert_frame_equal(sc.snps, self.snps_GRCh37())
assert sc.source == "generic, generic"