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_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_individual(self):
''' test that open_individual() works correctly
'''
# missing individual returns empty list
self.assertEqual(open_individual(None), [])
vcf = make_vcf_header()
vcf.append(make_vcf_line(pos=1, extra='HGNC=TEST;MAX_AF=0.0001'))
vcf.append(make_vcf_line(pos=2, extra='HGNC=ATRX;MAX_AF=0.0001'))
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
person = Person('fam_id', 'sample', 'dad', 'mom', 'F', '2', path)
var1 = SNV(chrom="1", position=1, id=".", ref="G", alts="T",
qual='1000', filter="PASS", info="CQ=missense_variant;HGNC=TEST;MAX_AF=0.0001",
format="DP:GT", sample="50:0/1", gender="female", mnv_code=None)
var2 = SNV(chrom="1", position=2, id=".", ref="G", alts="T",
qual='1000', filter="PASS", info="CQ=missense_variant;HGNC=ATRX;MAX_AF=0.0001",
format="DP:GT", sample="50:0/1", gender="female", mnv_code=None)
self.assertEqual(open_individual(person), [var2])
# define a set of variants to automatically pass, and check that these
# variants pass.
child_keys = set([('1', 1), ('1', 2)])
self.assertEqual(open_individual(person,
child_variants=child_keys), [var1, var2])
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_individual(self):
''' test that open_individual() works correctly
'''
# missing individual returns empty list
self.assertEqual(open_individual(None), [])
vcf = make_vcf_header()
vcf.append(make_vcf_line(pos=1, extra='HGNC=TEST;MAX_AF=0.0001'))
vcf.append(make_vcf_line(pos=2, extra='HGNC=ATRX;MAX_AF=0.0001'))
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
person = Person('fam_id', 'sample', 'dad', 'mom', 'F', '2', path)
var1 = SNV(chrom="1", position=1, id=".", ref="G", alts="T",
qual='1000', filter="PASS", info="CQ=missense_variant;HGNC=TEST;MAX_AF=0.0001",
format="DP:GT:AD", sample="50:0/1:10,10", gender="female", mnv_code=None)
var2 = SNV(chrom="1", position=2, id=".", ref="G", alts="T",
qual='1000', filter="PASS", info="CQ=missense_variant;HGNC=ATRX;MAX_AF=0.0001",
format="DP:GT:AD", sample="50:0/1:10,10", gender="female", mnv_code=None)
self.assertEqual(open_individual(person), [var2])
# define a set of variants to automatically pass, and check that these
# variants pass.
child_keys = set([('1', 1), ('1', 2)])
self.assertEqual(open_individual(person,
child_variants=child_keys), [var1, var2])
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_open_individual_male_het_chrx(self):
""" test that open_individual() passes over hets in males on chrX
"""
# the sub-functions are all tested elsewhere, this test merely checks
# that valid variants are added to the variants list, and invalid
# variants are passed over without being added to the variants list
vcf = make_vcf_header()
vcf.append(make_vcf_line(chrom='X', pos=1, genotype='0/1',
extra='HGNC=TEST;MAX_AF=0.0001'))
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
person = Person('fam_id', 'sample', 'dad', 'mom', 'M', '2', path)
self.assertEqual(open_individual(person), [])
def test_open_individual_male_het_chrx(self):
""" test that open_individual() passes over hets in males on chrX
"""
# the sub-functions are all tested elsewhere, this test merely checks
# that valid variants are added to the variants list, and invalid
# variants are passed over without being added to the variants list
vcf = make_vcf_header()
vcf.append(make_vcf_line(chrom='X', pos=1, genotype='0/1',
extra='HGNC=TEST;MAX_AF=0.0001'))
path = os.path.join(self.temp_dir, "temp.vcf")
write_temp_vcf(path, vcf)
person = Person('fam_id', 'sample', 'dad', 'mom', 'M', '2', path)
self.assertEqual(open_individual(person), [])