def test_export_gen_exprs(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
contig_recoding={
"01": "1"
},
reference_genome='GRCh37',
min_partitions=3).add_col_index().add_row_index()
out1 = new_temp_file()
hl.export_gen(gen,
out1,
id1=hl.str(gen.col_idx),
id2=hl.str(gen.col_idx),
missing=0.5,
varid=hl.str(gen.row_idx),
rsid=hl.str(gen.row_idx),
gp=[0.0, 1.0, 0.0])
in1 = (hl.import_gen(out1 + '.gen',
sample_file=out1 + '.sample',
min_partitions=3).add_col_index().add_row_index())
self.assertTrue(
in1.aggregate_entries(hl.agg.fraction(
in1.GP == [0.0, 1.0, 0.0])) == 1.0)
self.assertTrue(
in1.aggregate_rows(
hl.agg.fraction((in1.varid == hl.str(in1.row_idx))
& (in1.rsid == hl.str(in1.row_idx)))) == 1.0)
self.assertTrue(
in1.aggregate_cols(hl.agg.fraction(
(in1.s == hl.str(in1.col_idx)))))
def test_import_gen_skip_invalid_loci(self):
mt = hl.import_gen(resource('skip_invalid_loci.gen'),
resource('skip_invalid_loci.sample'),
reference_genome='GRCh37',
skip_invalid_loci=True)
self.assertTrue(mt._force_count_rows() == 3)
with self.assertRaisesRegex(FatalError, 'Invalid locus'):
hl.import_gen(resource('skip_invalid_loci.gen'),
resource('skip_invalid_loci.sample'))
def test_import_gen_skip_invalid_loci(self):
mt = hl.import_gen(resource('skip_invalid_loci.gen'),
resource('skip_invalid_loci.sample'),
reference_genome='GRCh37',
skip_invalid_loci=True)
self.assertTrue(mt._force_count_rows() == 3)
with self.assertRaisesRegex(FatalError, 'Invalid locus'):
hl.import_gen(resource('skip_invalid_loci.gen'),
resource('skip_invalid_loci.sample'))
def test_export_gen(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
contig_recoding={"01": "1"},
reference_genome='GRCh37',
min_partitions=3)
file = '/tmp/test_export_gen'
hl.export_gen(gen, file)
gen2 = hl.import_gen(file + '.gen',
sample_file=file + '.sample',
reference_genome='GRCh37',
min_partitions=3)
self.assertTrue(gen._same(gen2, tolerance=3E-4, absolute=True))
def test_import_gen_no_reference_specified(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
reference_genome=None)
self.assertTrue(gen.locus.dtype == hl.tstruct(contig=hl.tstr, position=hl.tint32))
self.assertEqual(gen.count_rows(), 199)
def test_import_gen_no_reference_specified(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
reference_genome=None)
self.assertTrue(gen.locus.dtype == hl.tstruct(contig=hl.tstr, position=hl.tint32))
self.assertEqual(gen.count_rows(), 199)
def test_import_gen(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
contig_recoding={"01": "1"},
reference_genome = 'GRCh37').rows()
self.assertTrue(gen.all(gen.locus.contig == "1"))
self.assertEqual(gen.count(), 199)
self.assertEqual(gen.locus.dtype, hl.tlocus('GRCh37'))
def test_import_gen(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
contig_recoding={"01": "1"},
reference_genome = 'GRCh37').rows()
self.assertTrue(gen.all(gen.locus.contig == "1"))
self.assertEqual(gen.count(), 199)
self.assertEqual(gen.locus.dtype, hl.tlocus('GRCh37'))
def test_import_bgen_random(self):
sample_file = resource('random.sample')
genmt = hl.import_gen(resource('random.gen'), sample_file)
bgen_file = resource('random.bgen')
hl.index_bgen(bgen_file)
bgenmt = hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file)
self.assertTrue(bgenmt._same(genmt, tolerance=1.0 / 255,
absolute=True))
def test_export_gen_exprs(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
contig_recoding={"01": "1"},
reference_genome='GRCh37',
min_partitions=3).add_col_index().add_row_index()
out1 = new_temp_file()
hl.export_gen(gen, out1, id1=hl.str(gen.col_idx), id2=hl.str(gen.col_idx), missing=0.5,
varid=hl.str(gen.row_idx), rsid=hl.str(gen.row_idx), gp=[0.0, 1.0, 0.0])
in1 = (hl.import_gen(out1 + '.gen', sample_file=out1 + '.sample', min_partitions=3)
.add_col_index()
.add_row_index())
self.assertTrue(in1.aggregate_entries(hl.agg.fraction(in1.GP == [0.0, 1.0, 0.0])) == 1.0)
self.assertTrue(in1.aggregate_rows(hl.agg.fraction((in1.varid == hl.str(in1.row_idx)) &
(in1.rsid == hl.str(in1.row_idx)))) == 1.0)
self.assertTrue(in1.aggregate_cols(hl.agg.fraction((in1.s == hl.str(in1.col_idx)))))
def test_multiple_files(self):
sample_file = resource('random.sample')
genmt = hl.import_gen(resource('random.gen'), sample_file)
bgen_file = [resource('random-b.bgen'), resource('random-c.bgen'), resource('random-a.bgen')]
hl.index_bgen(bgen_file)
bgenmt = hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file, n_partitions=3)
self.assertTrue(
bgenmt._same(genmt, tolerance=1.0 / 255, absolute=True))
def test_import_bgen_random(self):
sample_file = resource('random.sample')
genmt = hl.import_gen(resource('random.gen'), sample_file)
bgen_file = resource('random.bgen')
hl.index_bgen(bgen_file)
bgenmt = hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file)
self.assertTrue(
bgenmt._same(genmt, tolerance=1.0 / 255, absolute=True))
def test_export_gen(self):
gen = hl.import_gen(resource('example.gen'),
sample_file=resource('example.sample'),
contig_recoding={"01": "1"},
reference_genome='GRCh37',
min_partitions=3)
# permute columns so not in alphabetical order!
import random
indices = list(range(gen.count_cols()))
random.shuffle(indices)
gen = gen.choose_cols(indices)
file = '/tmp/test_export_gen'
hl.export_gen(gen, file)
gen2 = hl.import_gen(file + '.gen',
sample_file=file + '.sample',
reference_genome='GRCh37',
min_partitions=3)
self.assertTrue(gen._same(gen2, tolerance=3E-4, absolute=True))
def test_import_bgen_gavin_example(self):
recoding = {'0{}'.format(i): str(i) for i in range(1, 10)}
sample_file = resource('example.sample')
genmt = hl.import_gen(resource('example.gen'), sample_file,
contig_recoding=recoding,
reference_genome="GRCh37")
bgen_file = resource('example.8bits.bgen')
hl.index_bgen(bgen_file, contig_recoding=recoding,
reference_genome="GRCh37")
bgenmt = hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file)
self.assertTrue(
bgenmt._same(genmt, tolerance=1.0 / 255, absolute=True))
def test_multiple_files(self):
sample_file = resource('random.sample')
genmt = hl.import_gen(resource('random.gen'), sample_file)
bgen_file = [
resource('random-b.bgen'),
resource('random-c.bgen'),
resource('random-a.bgen')
]
hl.index_bgen(bgen_file)
bgenmt = hl.import_bgen(bgen_file, ['GT', 'GP'],
sample_file,
n_partitions=3)
self.assertTrue(bgenmt._same(genmt, tolerance=1.0 / 255,
absolute=True))
def test_import_bgen_gavin_example(self):
recoding = {'0{}'.format(i): str(i) for i in range(1, 10)}
sample_file = resource('example.sample')
genmt = hl.import_gen(resource('example.gen'),
sample_file,
contig_recoding=recoding)
bgen_file = resource('example.8bits.bgen')
hl.index_bgen(bgen_file)
bgenmt = hl.import_bgen(bgen_file, ['GT', 'GP'],
sample_file,
contig_recoding=recoding)
self.assertTrue(bgenmt._same(genmt, tolerance=1.0 / 255,
absolute=True))
