def test_fetch_bed():
import os
filenames = [
"chrom22_subsample20_maf0.10.bed",
"chrom22_subsample20_maf0.10.fam",
"chrom22_subsample20_maf0.10.bim",
]
with limix.file_example(filenames) as filepaths:
folder = os.path.dirname(filepaths[0])
filepath = os.path.join(folder, "chrom22_subsample20_maf0.10")
specs = [
f"{filepath}",
f"{filepath}:bed:row=candidate,col=sample",
f"{filepath}:bed:col=sample",
f"{filepath}:bed",
f"{filepath}",
f"{filepath}:",
f"{filepath}::",
f"{filepath}:bed:",
f"{filepath}::row=candidate",
]
for spec in specs:
G = fetch("genotype", spec, verbose=False)
assert_string_equal(G.name, "genotype")
assert_array_equal(G["sample"], _samples)
assert_equal(G.shape, (274, 49008))
assert_array_equal(G.dims, ["sample", "candidate"])
def test_fetch_csv():
with limix.file_example("expr.csv") as filepath:
spec = f"{filepath}:csv:row=trait,trait[gene1]"
y = fetch("trait", spec, verbose=False)
assert_string_equal(y.name, "trait")
assert_array_equal(y["sample"], _samples)
assert_equal(y.shape, (274, 1))
assert_allclose(y.values[:2, 0],
[-3.752_345_147_31, -0.421_128_991_488])
assert_array_equal(y.coords, ["trait", "sample"])
spec = f"{filepath}:csv:row=trait,trait[gene11]"
y = fetch("trait", spec, verbose=False)
assert_string_equal(y.name, "trait")
assert_array_equal(y["sample"], _samples)
assert_equal(y.shape, (274, 1))
assert_allclose(y.values[:2, 0], [0.798_312_717_19, 0.237_496_587_19])
spec = f"{filepath}:csv:row=trait"
y = fetch("trait", spec, verbose=False)
assert_string_equal(y.name, "trait")
assert_array_equal(y["sample"], _samples)
assert_equal(y.shape, (274, 11))
spec = f"{filepath}:csv:row=trait,col=sample"
y = fetch("trait", spec, verbose=False)
assert_equal(y.shape, (274, 11))
assert_equal(y.dims, ("sample", "trait"))
spec = f"{filepath}:csv:row=sample,col=trait"
y = fetch("trait", spec, verbose=False)
assert_equal(y.shape, (11, 274))
assert_equal(y.dims, ("sample", "trait"))
spec = f"{filepath}:csv:"
y = fetch("trait", spec, verbose=False)
spec = f"{filepath}:csv"
y = fetch("trait", spec, verbose=False)
spec = f"{filepath}:csv:row=samples"
with pytest.raises(ValueError):
y = fetch("trait", spec, verbose=False)
spec = "wrong_filepath:csv:row=sample"
with pytest.raises(FileNotFoundError):
y = fetch("trait", spec, verbose=False)
spec = f"{filepath}:csv:row=sample,col=trait"
with pytest.raises(ValueError):
y = fetch("traits", spec, verbose=False)
spec = f"{filepath}:csvs:row=sample,col=trait"
with pytest.raises(ValueError):
y = fetch("trait", spec, verbose=False)
def test_io_gen():
with file_example(["example.gen", "example.sample"]) as filepaths:
df = io.gen.read(filepaths[0][:-4], verbose=False)
assert_equal(df["sample"]["sample_id"][1], "1A1")
assert_equal(df["sample"]["age"][0], 4)
assert_allclose(df["genotype"]["1A4"]["AB"][0], 0.0207)
def test_io_hdf5():
with file_example("data.h5.bz2") as filepath:
filepath = extract(filepath, verbose=False)
with io.hdf5.fetcher(filepath) as df:
X = df.fetch("/genotype/chrom20/SNP")
assert_allclose(X[0, 0], [1.0])
def test_io_csv():
with file_example("pheno.csv") as filepath:
assert_equal(io.csv.read(filepath, verbose=False)["attr1"][0], "string")