def variant(self, case_id, variant_id):
"""Return a specific variant.
Args:
case_id (str): Path to vcf file
variant_id (str): A variant id
Returns:
variant (Variant): The variant object for the given id
"""
case_obj = self.case(case_id=case_id)
vcf_file_path = case_obj.variant_source
self.head = get_header(vcf_file_path)
self.vep_header = self.head.vep_columns
self.snpeff_header = self.head.snpeff_columns
handle = VCF(vcf_file_path)
for index, variant in enumerate(handle):
index += 1
line_id = get_variant_id(
variant_line=str(variant)).lstrip('chrCHR')
if line_id == variant_id:
return self._format_variants(variant=variant,
index=index,
case_obj=case_obj,
add_all_info=True)
return None
def variant(self, case_id, variant_id):
"""Return a specific variant.
Args:
case_id (str): Path to vcf file
variant_id (str): A variant id
Returns:
variant (Variant): The variant object for the given id
"""
case_obj = self.case(case_id=case_id)
vcf_file_path = case_obj.variant_source
head = self._get_header(vcf_file_path)
handle = get_vcf_handle(infile=vcf_file_path)
relevant_lines = (line for line in handle if not line.startswith('#'))
for index, variant_line in enumerate(relevant_lines):
index += 1
line_id = get_variant_id(variant_line=variant_line).lstrip('chrCHR')
if line_id == variant_id:
return self._format_variant(
variant_line=variant_line,
index=index,
case_obj=case_obj,
head=head
)
return None
def test_get_variant_id_dict():
"""docstring for test_get_variant_id_dict"""
variant_dict = {
'CHROM': '1',
'POS': '10',
'REF':'A',
'ALT': 'T'
}
assert get_variant_id(variant_dict) == '1_10_A_T'
def father(ctx):
"""Check number of variants in common"""
logger.info("Running variant_integrity father version {0}".format(
variant_integrity.__version__))
print_columns = ['ind_id', 'fraction_of_common_variants', 'common_variants', 'number_calls']
# Children is a dictionary of children that counts the number of errors
duos = []
children = {}
analysis_individuals = set()
for ind_id in ctx.parent.individuals:
individual_object = ctx.parent.individuals[ind_id]
if individual_object.father != '0':
duo = {
'child': ind_id,
'father': individual_object.father
}
analysis_individuals.add(ind_id)
analysis_individuals.add(individual_object.father)
duos.append(duo)
logger.info("Duo found: {0}".format(
', '.join(list(duo.values()))
))
logger.info("Individuals included in analysis: {0}".format(
','.join(list(analysis_individuals))))
for duo in duos:
children[duo['child']] = dict(zip(
print_columns, [duo['child'], 0, 0, 0]))
for line in ctx.parent.variant_file:
variant_dict = get_variant_dict(
variant_line=line,
header_line=ctx.parent.header_line
)
logger.debug("Checking genotype calls for variant {0}".format(
get_variant_id(variant_dict=variant_dict)
))
genotypes = get_genotypes(variant_dict, analysis_individuals)
for duo in duos:
child_id = duo['child']
child_genotype = genotypes[child_id]
father_genotype = genotypes[duo['father']]
duo_genotypes = [
child_genotype,
father_genotype
]
#First check if the child has the variant:
if child_genotype.has_variant:
# If child have high quality we count the variant
if check_high_quality([child_genotype], ctx.parent.gq_treshold):
children[child_id]['number_calls'] += 1
if check_common_variant(duo_genotypes):
children[child_id]['common_variants'] += 1
results = []
for child_id in children:
child_info = children[child_id]
common = child_info['common_variants']
variants = child_info['number_calls']
percentage = common/variants
child_info['fraction_of_common_variants'] = round(percentage, 3)
results.append(child_info)
to_json = ctx.parent.to_json
outfile = ctx.parent.outfile
if to_json:
if outfile:
json.dump(results, outfile)
else:
print(json.dumps(results))
else:
if outfile:
outfile.write("#{0}\n".format('\t'.join(print_columns)))
else:
print("#{0}".format('\t'.join(print_columns)))
for result in results:
print_line = "{0}\t{1}\t{2}\t{3}".format(
result['ind_id'], result['fraction_of_common_variants'],
result['common_variants'], result['number_calls']
)
if outfile:
outfile.write("{0}\n".format(print_line))
else:
print(print_line)
def mendel(ctx):
"""Check mendelian errors in all trios"""
logger.info("Running variant_integrity mendel {0}".format(
variant_integrity.__version__))
print_columns = ['ind_id', 'fraction_of_errors', 'mendelian_errors', 'number_calls']
# Children is a dictionary of children that counts the number of errors
trios = []
children = {}
analysis_individuals = set()
for family in ctx.parent.families:
family_object = ctx.parent.families[family]
for trio in family_object.trios:
trio_individuals = {
'mother':None,
'father':None,
'child':None,
}
for ind_id in trio:
analysis_individuals.add(ind_id)
individual_object = ctx.parent.individuals[ind_id]
if individual_object.mother in trio:
trio_individuals['child'] = ind_id
elif individual_object.sex == 1:
trio_individuals['father'] = ind_id
else:
trio_individuals['mother'] = ind_id
trios.append(trio_individuals)
logger.info("Trio found: {0}".format(
', '.join(list(trio_individuals.values()))
))
logger.info("Individuals included in analysis: {0}".format(
','.join(list(analysis_individuals))))
for trio in trios:
children[trio['child']] = dict(zip(
print_columns, [trio['child'], 0, 0, 0]))
for line in ctx.parent.variant_file:
variant_dict = get_variant_dict(
variant_line=line,
header_line=ctx.parent.header_line
)
logger.debug("Checking genotype calls for variant {0}".format(
get_variant_id(variant_dict=variant_dict)
))
genotypes = get_genotypes(variant_dict, analysis_individuals)
for trio in trios:
child_id = trio['child']
child_genotype = genotypes[child_id]
mother_genotype = genotypes[trio['mother']]
father_genotype = genotypes[trio['father']]
trio_genotypes = [
child_genotype,
mother_genotype,
father_genotype
]
#First check if the child has the variant:
if child_genotype.has_variant:
# If all individuals are high quality we count the variant
if check_high_quality(trio_genotypes, ctx.parent.gq_treshold):
children[child_id]['number_calls'] += 1
if check_mendelian_error(child_genotype, mother_genotype, father_genotype):
children[child_id]['mendelian_errors'] += 1
results = []
for child_id in children:
child_info = children[child_id]
errors = child_info['mendelian_errors']
variants = child_info['number_calls']
percentage = errors/variants
child_info['fraction_of_errors'] = round(percentage, 3)
results.append(child_info)
to_json = ctx.parent.to_json
outfile = ctx.parent.outfile
if to_json:
if outfile:
json.dump(results, outfile)
else:
print(json.dumps(results))
else:
if outfile:
outfile.write("#{0}\n".format('\t'.join(print_columns)))
else:
print("#{0}".format('\t'.join(print_columns)))
for result in results:
print_line = "{0}\t{1}\t{2}\t{3}".format(
result['ind_id'], result['fraction_of_errors'],
result['mendelian_errors'], result['number_calls']
)
if outfile:
outfile.write("{0}\n".format(print_line))
else:
print(print_line)
def test_send_wrong():
"""docstring for test_get_variant_id_dict"""
with pytest.raises(Exception):
get_variant_id()
def test_get_variant_id_line():
"""docstring for test_get_variant_id_dict"""
variant_line = '1\t10\t.\tA\tT'
assert get_variant_id(variant_line=variant_line) == '1_10_A_T'
def test_get_variant_id_dict():
"""docstring for test_get_variant_id_dict"""
variant_dict = {'CHROM': '1', 'POS': '10', 'REF': 'A', 'ALT': 'T'}
assert get_variant_id(variant_dict) == '1_10_A_T'
def test_send_wrong():
"""docstring for test_get_variant_id_dict"""
with pytest.raises(Exception):
get_variant_id()
def test_get_variant_id_line():
"""docstring for test_get_variant_id_dict"""
variant_line = '1\t10\t.\tA\tT'
assert get_variant_id(variant_line=variant_line) == '1_10_A_T'
