def __iter__(self):
if not self.metadata.fileformat:
raise SyntaxError("Vcf must have fileformat defined")
if self.vcf:
# We need to treat the first case as an exception
if self.beginning:
variants = []
if self.next_line:
first_variant = format_variant(line=self.next_line,
header_parser=self.metadata,
check_info=self.check_info)
if not (self.split_variants
and len(first_variant['ALT'].split(',')) > 1):
variants.append(first_variant)
else:
for splitted_variant in split_variants(
variant_dict=first_variant,
header_parser=self.metadata,
allele_symbol=self.allele_symbol):
variants.append(splitted_variant)
for variant in variants:
yield variant
self.beginning = False
for line in self.vcf:
line = line.rstrip()
# These are the variant(s) found in one line of the vcf
# If there are multiple alternatives and self.split_variants
# There can be more than one variant in one line
variants = []
if not line.startswith('#') and len(line.split('\t')) >= 8:
variant = format_variant(line=line,
header_parser=self.metadata,
check_info=self.check_info)
if not (self.split_variants
and len(variant['ALT'].split(',')) > 1):
variants.append(variant)
else:
for splitted_variant in split_variants(
variant_dict=variant,
header_parser=self.metadata,
allele_symbol=self.allele_symbol):
variants.append(splitted_variant)
for variant in variants:
yield variant
else:
for variant in self.variants:
yield variant
def add_variant(self, chrom, pos, rs_id, ref, alt, qual, filt, info, form=None, genotypes=[]):
"""
Add a variant to the parser.
This function is for building a vcf. It takes the relevant parameters
and make a vcf variant in the proper format.
"""
variant_info = [chrom, pos, rs_id, ref, alt, qual, filt, info]
if form:
variant_info.append(form)
for individual in genotypes:
variant_info.append(individual)
variant_line = '\t'.join(variant_info)
variant = format_variant(
line = variant_line,
header_parser = self.metadata,
check_info = self.check_info
)
if not (self.split_variants and len(variant['ALT'].split(',')) > 1):
self.variants.append(variant)
# If multiple alternative and split_variants we must split the variant
else:
for splitted_variant in split_variants(
variant_dict=variant,
header_parser=self.metadata,
allele_symbol=self.allele_symbol):
self.variants.append(splitted_variant)
def test_simple_variant():
"""
Test how the format_variant behaves
"""
header_parser = get_header()
variant_line = "1\t11900\t.\tA\tT\t100\tPASS\tMQ=1\tGT:GQ\t0/1:60\t"\
"0/1:60\t1/1:60"
variant = format_variant(line=variant_line,
header_parser=header_parser,
check_info=True)
info_dict = OrderedDict()
info_dict['MQ'] = ['1']
assert variant['CHROM'] == "1"
assert variant['POS'] == "11900"
assert variant['ID'] == "."
assert variant['REF'] == "A"
assert variant['ALT'] == "T"
assert variant['QUAL'] == "100"
assert variant['FILTER'] == "PASS"
assert variant['INFO'] == "MQ=1"
assert variant['FORMAT'] == "GT:GQ"
assert variant['father'] == "0/1:60"
assert variant['mother'] == "0/1:60"
assert variant['proband'] == "1/1:60"
assert variant['info_dict'] == info_dict
assert type(variant['genotypes']['mother']) == type(Genotype())
def test_simple_variant():
"""
Test how the format_variant behaves
"""
header_parser = get_header()
variant_line = "1\t11900\t.\tA\tT\t100\tPASS\tMQ=1\tGT:GQ\t0/1:60\t"\
"0/1:60\t1/1:60"
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
info_dict = OrderedDict()
info_dict['MQ'] = ['1']
assert variant['CHROM'] == "1"
assert variant['POS'] == "11900"
assert variant['ID'] == "."
assert variant['REF'] == "A"
assert variant['ALT'] == "T"
assert variant['QUAL'] == "100"
assert variant['FILTER'] == "PASS"
assert variant['INFO'] == "MQ=1"
assert variant['FORMAT'] == "GT:GQ"
assert variant['father'] == "0/1:60"
assert variant['mother'] == "0/1:60"
assert variant['proband'] == "1/1:60"
assert variant['info_dict'] == info_dict
assert type(variant['genotypes']['mother']) == type(Genotype())
def test_split_minimal():
"""
Test to split a vcf line without genotypes
"""
header_lines = [
'##fileformat=VCFv4.2',
'##FILTER=<ID=LowQual,Description="Low quality">',
'##INFO=<ID=MQ,Number=1,Type=Float,Description="RMS Mapping Quality">',
'##contig=<ID=1,length=249250621,assembly=b37>',
'#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO'
]
header_parser = get_header(header_lines)
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tMQ=1"
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
splitted_variants = []
for variant in split_variants(variant, header_parser):
splitted_variants.append(variant)
assert len(splitted_variants) == 2
def test_split_no_info():
"""
Test how split genotypes when wrong number of entrys
"""
header_parser = get_header()
# CNT should have two entrys since Number=A
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\t."\
"\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
#But then we need to skip the info check
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=False
)
splitted_variants = []
for variant in split_variants(variant, header_parser):
splitted_variants.append(variant)
assert len(splitted_variants) == 2
first_variant = splitted_variants[0]
second_variant = splitted_variants[1]
assert first_variant['info_dict'] == {'.':[]}
assert second_variant['info_dict'] == {'.':[]}
assert first_variant['INFO'] == '.'
assert second_variant['INFO'] == '.'
def test_wrong_number_of_R_entrys():
"""
Test how split genotypes when wrong number of entrys
"""
header_parser = get_header()
# CNT should have two entrys since Number=A
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tMQ=1;CNT=5,8;"\
"DP_HIST=12,43\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
#But then we need to skip the info check
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=False
)
splitted_variants = []
for variant in split_variants(variant, header_parser):
splitted_variants.append(variant)
assert len(splitted_variants) == 2
first_variant = splitted_variants[0]
second_variant = splitted_variants[1]
#Vcf-parser should use the first annotation for both alleles
assert first_variant['info_dict']['DP_HIST'] == ['12','43']
assert second_variant['info_dict']['DP_HIST'] == ['12','43']
def test_simple_split():
"""
Test how split genotypes behave when a simple split
"""
header_parser = get_header()
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tMQ=1;CNT=5,8;"\
"DP_HIST=12,43,22\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
variant = variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
splitted_variants = []
for variant in split_variants(variant, header_parser):
splitted_variants.append(variant)
assert len(splitted_variants) == 2
first_variant = splitted_variants[0]
second_variant = splitted_variants[1]
# Test if the splitted variants still have the same reference
assert first_variant['REF'] == 'A'
assert second_variant['REF'] == 'A'
# Test if the alternative was splitted properly
assert first_variant['ALT'] == 'T'
assert second_variant['ALT'] == 'C'
# Test if simple ino field is handled correct
assert first_variant['info_dict']['MQ'] == ['1']
assert second_variant['info_dict']['MQ'] == ['1']
# Test if info field with Number='A' is handled correct
assert first_variant['info_dict']['CNT'] == ['5']
assert second_variant['info_dict']['CNT'] == ['8']
# Test if info field with Number='R' is handled correct
assert first_variant['info_dict']['DP_HIST'] == ['12', '43']
assert second_variant['info_dict']['DP_HIST'] == ['12', '22']
# Test if the genortypes are on the correct format
assert first_variant['father'] == "1/1:60:0,7:12"
assert second_variant['father'] == "0/0:60:0,0:12"
assert first_variant['mother'] == "0/0:60:7,0:17"
assert second_variant['mother'] == "0/1:60:7,10:17"
assert first_variant['proband'] == "0/1:60:0,7:16"
assert second_variant['proband'] == "0/1:60:0,8:16"
def test_malformed_line():
"""
Test if proper behaviour with malformed vcf line
"""
header_parser = get_header()
# Missing position
variant_line = "1\t.\tA\tT\t100\tPASS\tMQ=1\tGT:GQ\t0/1:60\t"\
"0/1:60\t1/1:60"
with pytest.raises(SyntaxError):
variant = format_variant(line=variant_line,
header_parser=header_parser,
check_info=True)
def test_wrong_number_annotation_genotype():
"""
Test if proper behaviour with malformed vcf line
"""
header_parser = get_header()
# Missing position
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tMQ=1;CNT=5,6;"\
"SQ=1,2\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
with pytest.raises(SyntaxError):
variant = format_variant(line=variant_line,
header_parser=header_parser,
check_info=True)
def test_malformed_line():
"""
Test if proper behaviour with malformed vcf line
"""
header_parser = get_header()
# Missing position
variant_line = "1\t.\tA\tT\t100\tPASS\tMQ=1\tGT:GQ\t0/1:60\t"\
"0/1:60\t1/1:60"
with pytest.raises(SyntaxError):
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
def test_wrong_number_annotation_genotype():
"""
Test if proper behaviour with malformed vcf line
"""
header_parser = get_header()
# Missing position
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tMQ=1;CNT=5,6;"\
"SQ=1,2\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
with pytest.raises(SyntaxError):
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
def test_no_genotypes():
"""
Test if proper behaviour with minimal vcf
"""
header_lines = [
'##fileformat=VCFv4.2',
'##FILTER=<ID=LowQual,Description="Low quality">',
'##INFO=<ID=MQ,Number=1,Type=Float,Description="RMS Mapping Quality">',
'##contig=<ID=1,length=249250621,assembly=b37>',
'#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO'
]
header_parser = get_header(header_lines)
# Missing position
variant_line = "1\t11900\t.\tA\tT\t100\tPASS\tMQ=1"
variant = format_variant(line=variant_line,
header_parser=header_parser,
check_info=True)
def test_no_genotypes():
"""
Test if proper behaviour with minimal vcf
"""
header_lines = [
'##fileformat=VCFv4.2',
'##FILTER=<ID=LowQual,Description="Low quality">',
'##INFO=<ID=MQ,Number=1,Type=Float,Description="RMS Mapping Quality">',
'##contig=<ID=1,length=249250621,assembly=b37>',
'#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO'
]
header_parser = get_header(header_lines)
# Missing position
variant_line = "1\t11900\t.\tA\tT\t100\tPASS\tMQ=1"
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
def test_csq_split():
"""
Test works when splitting CSQ fields
"""
header_parser = get_header()
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tCSQ=T|148398|NM_152486.2,"\
"C|148398|NM_152486.2\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
splitted_variants = []
for variant in split_variants(variant, header_parser):
splitted_variants.append(variant)
assert len(splitted_variants) == 2
first_variant = splitted_variants[0]
second_variant = splitted_variants[1]
assert first_variant['info_dict']['CSQ'] == ['T|148398|NM_152486.2']
assert second_variant['info_dict']['CSQ'] == ['C|148398|NM_152486.2']
assert list(first_variant['vep_info'].keys()) == ['T']
assert list(second_variant['vep_info'].keys()) == ['C']
assert first_variant['vep_info']['T'] == [{
'Allele':'T',
'Gene':'148398',
'Feature':'NM_152486.2'
}]
def test_csq_split_missing_allele():
"""
Test works when splitting CSQ fields where one allele is missing
"""
header_parser = get_header()
variant_line = "3\t947379\t.\tA\tT,C\t100\tPASS\tCSQ=T|148398|NM_152486.2"\
"\tGT:GQ:AD:DP\t1/1:60:0,7,0:12\t0/2:60:7,0,10:17"\
"\t1/2:60:0,7,8:16"
variant = format_variant(
line = variant_line,
header_parser=header_parser,
check_info=True
)
splitted_variants = []
for variant in split_variants(variant, header_parser):
splitted_variants.append(variant)
assert len(splitted_variants) == 2
first_variant = splitted_variants[0]
second_variant = splitted_variants[1]
assert first_variant['info_dict']['CSQ'] == ['T|148398|NM_152486.2']
with pytest.raises(KeyError):
assert second_variant['info_dict']['CSQ'] == ['']
assert list(first_variant['vep_info'].keys()) == ['T']
assert list(second_variant['vep_info'].keys()) == ['C']
assert second_variant['vep_info']['C'] == []
def __iter__(self):
if not self.metadata.fileformat:
raise SyntaxError("Vcf must have fileformat defined")
if self.vcf:
# We need to treat the first case as an exception
if self.beginning:
variants = []
first_variant = format_variant(
line = self.next_line,
header_parser = self.metadata,
check_info = self.check_info
)
if not (self.split_variants and len(first_variant['ALT'].split(',')) > 1):
variants.append(first_variant)
else:
for splitted_variant in split_variants(
variant_dict=first_variant,
header_parser=self.metadata,
allele_symbol=self.allele_symbol):
variants.append(splitted_variant)
for variant in variants:
yield variant
self.beginning = False
for line in self.vcf:
line = line.rstrip()
# These are the variant(s) found in one line of the vcf
# If there are multiple alternatives and self.split_variants
# There can be more than one variant in one line
variants = []
if not line.startswith('#') and len(line.split('\t')) >= 8:
variant = format_variant(
line = line,
header_parser = self.metadata,
check_info = self.check_info
)
if not (self.split_variants and len(variant['ALT'].split(',')) > 1):
variants.append(variant)
else:
for splitted_variant in split_variants(
variant_dict=variant,
header_parser=self.metadata,
allele_symbol=self.allele_symbol):
variants.append(splitted_variant)
for variant in variants:
yield variant
else:
for variant in self.variants:
yield variant