def test_import_table_force_bgz(self):
f = new_temp_file(suffix=".bgz")
t = hl.utils.range_table(10, 5)
t.export(f)
f2 = new_temp_file(suffix=".gz")
run_command(["cp", uri_path(f), uri_path(f2)])
t2 = hl.import_table(f2, force_bgz=True, impute=True).key_by('idx')
self.assertTrue(t._same(t2))
def test_import_table_force_bgz(self):
f = new_temp_file(suffix=".bgz")
t = hl.utils.range_table(10, 5)
t.export(f)
f2 = new_temp_file(suffix=".gz")
run_command(["cp", uri_path(f), uri_path(f2)])
t2 = hl.import_table(f2, force_bgz=True, impute=True).key_by('idx')
self.assertTrue(t._same(t2))
def test_export_plink(self):
vcf_file = resource('sample.vcf')
mt = hl.split_multi_hts(hl.import_vcf(vcf_file, min_partitions=10))
# permute columns so not in alphabetical order!
import random
indices = list(range(mt.count_cols()))
random.shuffle(indices)
mt = mt.choose_cols(indices)
split_vcf_file = uri_path(new_temp_file())
hl_output = uri_path(new_temp_file())
plink_output = uri_path(new_temp_file())
merge_output = uri_path(new_temp_file())
hl.export_vcf(mt, split_vcf_file)
hl.export_plink(mt, hl_output)
run_command(["plink", "--vcf", split_vcf_file,
"--make-bed", "--out", plink_output,
"--const-fid", "--keep-allele-order"])
data = []
with open(uri_path(plink_output + ".bim")) as file:
for line in file:
row = line.strip().split()
row[1] = ":".join([row[0], row[3], row[5], row[4]])
data.append("\t".join(row) + "\n")
with open(plink_output + ".bim", 'w') as f:
f.writelines(data)
run_command(["plink", "--bfile", plink_output,
"--bmerge", hl_output, "--merge-mode",
"6", "--out", merge_output])
same = True
with open(merge_output + ".diff") as f:
for line in f:
row = line.strip().split()
if row != ["SNP", "FID", "IID", "NEW", "OLD"]:
same = False
break
self.assertTrue(same)
def test_export_plink(self):
vcf_file = resource('sample.vcf')
mt = hl.split_multi_hts(hl.import_vcf(vcf_file, min_partitions=10))
split_vcf_file = uri_path(new_temp_file())
hl_output = uri_path(new_temp_file())
plink_output = uri_path(new_temp_file())
merge_output = uri_path(new_temp_file())
hl.export_vcf(mt, split_vcf_file)
hl.export_plink(mt, hl_output)
run_command([
"plink", "--vcf", split_vcf_file, "--make-bed", "--out",
plink_output, "--const-fid", "--keep-allele-order"
])
data = []
with open(uri_path(plink_output + ".bim")) as file:
for line in file:
row = line.strip().split()
row[1] = ":".join([row[0], row[3], row[5], row[4]])
data.append("\t".join(row) + "\n")
with open(plink_output + ".bim", 'w') as f:
f.writelines(data)
run_command([
"plink", "--bfile", plink_output, "--bmerge", hl_output,
"--merge-mode", "6", "--out", merge_output
])
same = True
with open(merge_output + ".diff") as f:
for line in f:
row = line.strip().split()
if row != ["SNP", "FID", "IID", "NEW", "OLD"]:
same = False
break
self.assertTrue(same)
def test_old_index_file_throws_error(self):
sample_file = resource('random.sample')
bgen_file = resource('random.bgen')
# missing file
if os.path.exists(bgen_file + '.idx2'):
run_command(['rm', '-r', bgen_file + '.idx2'])
with self.assertRaisesRegex(FatalError, 'have no .idx2 index file'):
hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file, n_partitions=3)
# old index file
run_command(['touch', bgen_file + '.idx'])
with self.assertRaisesRegex(FatalError, 'have no .idx2 index file'):
hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file)
run_command(['rm', bgen_file + '.idx'])
def test_old_index_file_throws_error(self):
sample_file = resource('random.sample')
bgen_file = resource('random.bgen')
# missing file
if os.path.exists(bgen_file + '.idx2'):
run_command(['rm', '-r', bgen_file + '.idx2'])
with self.assertRaisesRegex(FatalError, 'have no .idx2 index file'):
hl.import_bgen(bgen_file, ['GT', 'GP'],
sample_file,
n_partitions=3)
# old index file
run_command(['touch', bgen_file + '.idx'])
with self.assertRaisesRegex(FatalError, 'have no .idx2 index file'):
hl.import_bgen(bgen_file, ['GT', 'GP'], sample_file)
run_command(['rm', bgen_file + '.idx'])