def test_load_variants(self):
''' test that load_variants() works correctly. Mainly checks variables
are set correctly.
'''
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
sum_x_lr2 = {}
parents = True
fam = Family(
'fam', children=[Person('fam', 'child', '0', '0', 'f', '2', path)])
variants = load_variants(fam, 0.9, ['AFR_AF'], self.known_genes, set(),
sum_x_lr2, parents)
self.assertEqual(SNV.known_genes, self.known_genes)
self.assertEqual(CNV.known_genes, self.known_genes)
self.assertEqual(Info.populations, ['AFR_AF'])
self.assertEqual(Info.last_base, set())
# and check that the
variants = load_variants(fam, 0.9, [], None, set([('1', 100)]),
sum_x_lr2, parents)
self.assertIsNone(SNV.known_genes, self.known_genes)
self.assertIsNone(CNV.known_genes, self.known_genes)
self.assertEqual(Info.populations, [])
self.assertEqual(Info.last_base, set([('1', 100)]))
def test_open_vcf(self):
""" test obtaining a file handle for the VCF
"""
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
# check that plain VCF files can be loaded
handle = open_vcf(path)
self.assertEqual(type(handle), io.TextIOWrapper)
handle.close()
# check that gzipped vcf files are handled correctly
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
handle = open_vcf(path)
if IS_PYTHON3:
self.assertEqual(type(handle), io.TextIOWrapper)
else:
self.assertEqual(type(handle), gzip.GzipFile)
handle.close()
# make sure files that don't exists raise an error
path = os.path.join(self.temp_dir, "zzz.txt")
with self.assertRaises(OSError):
open_vcf(path)
# check that files with unknown extensions raise errors
path = os.path.join(self.temp_dir, "temp.zzz")
write_temp_vcf(path, vcf)
with self.assertRaises(OSError):
open_vcf(path)
def test_exclude_header(self):
""" test that exclude_header() works correctly
"""
vcf = make_minimal_vcf()
# make sure we drop the header, and only the header from the file
# check this by reading the file, and making sure the first line
# is the line we expect from the VCF
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
handler = open(path, "r")
exclude_header(handler)
self.assertEqual(handler.readline(), vcf[4])
handler.close()
# also check for gzipped VCF files.
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
mode = 'r'
if IS_PYTHON3:
mode = 'rt'
with gzip.open(path, mode) as handler:
exclude_header(handler)
self.assertEqual(handler.readline(), vcf[4])
def test_open_vcf(self):
""" test obtaining a file handle for the VCF
"""
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
# check that plain VCF files can be loaded
handle = open_vcf(path)
self.assertEqual(type(handle), io.TextIOWrapper)
handle.close()
# check that gzipped vcf files are handled correctly
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
handle = open_vcf(path)
if IS_PYTHON3:
self.assertEqual(type(handle), io.TextIOWrapper)
else:
self.assertEqual(type(handle), gzip.GzipFile)
handle.close()
# make sure files that don't exists raise an error
path = os.path.join(self.temp_dir, "zzz.txt")
with self.assertRaises(OSError):
open_vcf(path)
# check that files with unknown extensions raise errors
path = os.path.join(self.temp_dir, "temp.zzz")
write_temp_vcf(path, vcf)
with self.assertRaises(OSError):
open_vcf(path)
def test_debug_option(self):
""" test whether we can set up the class with the debug option
"""
known = {}
pops = None
sum_x_lr2 = {}
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
fam = Family(
'fam', children=[Person('fam', 'child', '0', '0', 'f', '2', path)])
# if the debug info isn't available, then the SNV object doesn't use the
# debug filter function
variants = load_variants(fam, 1.0, pops, known, set(), sum_x_lr2)
self.assertNotEqual(SNV.passes_filters, SNV.passes_filters_with_debug)
# if the debug info is passed in, check that the debug filter function
# got set correctly
variants = load_variants(fam, 1.0, pops, known, set(), sum_x_lr2, "1",
"10000")
self.assertEqual(SNV.passes_filters, SNV.passes_filters_with_debug)
def test_exclude_header(self):
""" test that exclude_header() works correctly
"""
vcf = make_minimal_vcf()
# make sure we drop the header, and only the header from the file
# check this by reading the file, and making sure the first line
# is the line we expect from the VCF
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
handler = open(path, "r")
exclude_header(handler)
self.assertEqual(handler.readline(), vcf[4])
handler.close()
# also check for gzipped VCF files.
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
mode = 'r'
if IS_PYTHON3:
mode = 'rt'
with gzip.open(path, mode) as handler:
exclude_header(handler)
self.assertEqual(handler.readline(), vcf[4])
def test_load_variants(self):
''' test that load_variants() works correctly. Mainly checks variables
are set correctly.
'''
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
sum_x_lr2 = {}
parents = True
fam = Family('fam', children=[Person('fam', 'child', '0', '0', 'f', '2', path)])
variants = load_variants(fam, 0.9, ['AFR_AF'], self.known_genes, set(), sum_x_lr2, parents)
self.assertEqual(SNV.known_genes, self.known_genes)
self.assertEqual(CNV.known_genes, self.known_genes)
self.assertEqual(Info.populations, ['AFR_AF'])
self.assertEqual(Info.last_base, set())
# and check that the
variants = load_variants(fam, 0.9, [], None, set([('1', 100)]), sum_x_lr2, parents)
self.assertIsNone(SNV.known_genes, self.known_genes)
self.assertIsNone(CNV.known_genes, self.known_genes)
self.assertEqual(Info.populations, [])
self.assertEqual(Info.last_base, set([('1', 100)]))
def test_get_vcf_header(self):
""" test that get_vcf_header() works correctly
"""
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
header = get_vcf_header(path)
# check that the header is returned correctly
self.assertEqual(header, vcf[:4])
def test_get_vcf_header(self):
""" test that get_vcf_header() works correctly
"""
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
header = get_vcf_header(path)
# check that the header is returned correctly
self.assertEqual(header, vcf[:4])
def test_get_vcf_provenance(self):
""" test that get_vcf_provenance() works correctly
"""
path = os.path.join(self.temp_dir, "temp.vcf")
gz_path = os.path.join(self.temp_dir, "temp.vcf.gz")
date_path = os.path.join(self.temp_dir, "temp.process.2014-02-20.vcf")
family = Family('famid')
family.add_child('child_id', 'mother', 'father', 'f', '2', path)
family.add_mother('mom_id', '0', '0', 'female', '1', gz_path)
family.add_father('dad_id', '0', '0', 'male', '1', date_path)
family.set_child()
vcf = make_minimal_vcf()
vcf_string = "".join(vcf)
if IS_PYTHON3:
vcf_string = vcf_string.encode("utf-8")
ungzipped_hash = hashlib.sha1(vcf_string).hexdigest()
header = vcf[:4]
write_temp_vcf(path, vcf)
# check that the file defs return correctly
(checksum, basename, date) = get_vcf_provenance(family.child)
self.assertEqual(checksum, ungzipped_hash)
self.assertEqual(basename, "temp.vcf")
self.assertEqual(date, "2014-01-01")
# now write a gzip file, and check that we get the correct hash
write_gzipped_vcf(gz_path, vcf)
handle = open(gz_path, "rb")
gzipped_hash = hashlib.sha1(handle.read()).hexdigest()
handle.close()
(checksum, basename, date) = get_vcf_provenance(family.mother)
self.assertEqual(checksum, gzipped_hash)
# check that when a fileDate isn't available in the VCF, we can pick
# the date from the path
vcf.pop(1)
write_temp_vcf(date_path, vcf)
(checksum, basename, date) = get_vcf_provenance(family.father)
self.assertEqual(date, "2014-02-20")
# and check we get null values if the family member is not present
family.father = None
provenance = get_vcf_provenance(family.father)
self.assertEqual(provenance, ('NA', 'NA', 'NA'))
def test_get_vcf_provenance(self):
""" test that get_vcf_provenance() works correctly
"""
path = os.path.join(self.temp_dir, "temp.vcf")
gz_path = os.path.join(self.temp_dir, "temp.vcf.gz")
date_path = os.path.join(self.temp_dir, "temp.process.2014-02-20.vcf")
family = Family('famid')
family.add_child('child_id', 'mother', 'father', 'f', '2', path)
family.add_mother('mom_id', '0', '0', 'female', '1', gz_path)
family.add_father('dad_id', '0', '0', 'male', '1', date_path)
family.set_child()
vcf = make_minimal_vcf()
vcf_string = "".join(vcf)
if IS_PYTHON3:
vcf_string = vcf_string.encode("utf-8")
ungzipped_hash = hashlib.sha1(vcf_string).hexdigest()
header = vcf[:4]
write_temp_vcf(path, vcf)
# check that the file defs return correctly
(checksum, basename, date) = get_vcf_provenance(family.child)
self.assertEqual(checksum, ungzipped_hash)
self.assertEqual(basename, "temp.vcf")
self.assertEqual(date, "2014-01-01")
# now write a gzip file, and check that we get the correct hash
write_gzipped_vcf(gz_path, vcf)
handle = open(gz_path, "rb")
gzipped_hash = hashlib.sha1(handle.read()).hexdigest()
handle.close()
(checksum, basename, date) = get_vcf_provenance(family.mother)
self.assertEqual(checksum, gzipped_hash)
# check that when a fileDate isn't available in the VCF, we can pick
# the date from the path
vcf.pop(1)
write_temp_vcf(date_path, vcf)
(checksum, basename, date) = get_vcf_provenance(family.father)
self.assertEqual(date, "2014-02-20")
# and check we get null values if the family member is not present
family.father = None
provenance = get_vcf_provenance(family.father)
self.assertEqual(provenance, ('NA', 'NA', 'NA'))
def test_debug_option(self):
""" test whether we can set up the class with the debug option
"""
known = {}
pops = None
sum_x_lr2 = {}
vcf = make_minimal_vcf()
path = os.path.join(self.temp_dir, "temp.vcf.gz")
write_gzipped_vcf(path, vcf)
fam = Family('fam', children=[Person('fam', 'child', '0', '0', 'f', '2', path)])
# if the debug info isn't available, then the SNV object doesn't use the
# debug filter function
variants = load_variants(fam, 1.0, pops, known, set(), sum_x_lr2)
self.assertNotEqual(SNV.passes_filters, SNV.passes_filters_with_debug)
# if the debug info is passed in, check that the debug filter function
# got set correctly
variants = load_variants(fam, 1.0, pops, known, set(), sum_x_lr2, "1", "10000")
self.assertEqual(SNV.passes_filters, SNV.passes_filters_with_debug)