def setUp(self):
self.converter = VcfToBedpeConverter()
header_lines = [
'##fileformat=VCFv4.2', '##fileDate=20090805',
'##reference=file:///seq/references/1000GenomesPilot-NCBI36.fasta',
'##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">',
'##INFO=<ID=AF,Number=A,Type=Float,Description="Allele Frequency">',
'##INFO=<ID=END,Number=1,Type=Integer,Description="End position of the variant">',
'##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">',
'#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT NA00001'
]
self.vcf = Vcf()
self.vcf.add_header(header_lines)
def setUp(self):
self.converter = VcfToBedpeConverter()
header_lines = [
'##fileformat=VCFv4.2',
'##fileDate=20090805',
'##reference=file:///seq/references/1000GenomesPilot-NCBI36.fasta',
'##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">',
'##INFO=<ID=AF,Number=A,Type=Float,Description="Allele Frequency">',
'##INFO=<ID=END,Number=1,Type=Integer,Description="End position of the variant">',
'##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">',
'#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT NA00001']
self.vcf = Vcf()
self.vcf.add_header(header_lines)
def vcfToBedpe(vcf_file, bedpe_out):
converter = VcfToBedpeConverter()
vcf = svtools.vcf.file.Vcf()
in_header = True
header = []
sample_list = []
bnds = dict()
sec_bnds = dict()
v = []
for line in vcf_file:
if in_header:
if line[0:2] == '##':
if line.split('=')[0] == '##fileformat':
line = '##fileformat=' + "BEDPE" + '\n'
if line.split('=')[0] == '##fileDate':
line = '##fileDate=' + time.strftime('%Y%m%d') + '\n'
header.append(line)
continue
elif line[0] == '#' and line[1] != '#':
sample_list = line.rstrip().split('\t')[9:]
header.append(line)
continue
else:
# print header
in_header = False
vcf.add_header(header)
if "SVTYPE" in [info.id for info in vcf.info_list]:
vcf.add_info_after("SVTYPE", "POS", 1, 'Integer', 'Position of the variant described in this record')
header=vcf.get_header()
bedpe_out.write(header[:header.rfind('\n')] + '\n')
final_header_line = ['#CHROM_A',
'START_A',
'END_A',
'CHROM_B',
'START_B',
'END_B',
'ID',
'QUAL',
'STRAND_A',
'STRAND_B',
'TYPE',
'FILTER',
'NAME_A',
'REF_A',
'ALT_A',
'NAME_B',
'REF_B',
'ALT_B',
'INFO_A',
'INFO_B']
if len(sample_list) > 0:
bedpe_out.write('\t'.join(final_header_line + ['FORMAT','\t'.join(map(str,sample_list))]) + '\n')
else:
bedpe_out.write('\t'.join(final_header_line) + '\n')
v = line.rstrip().split('\t')
var = svtools.vcf.variant.Variant(v, vcf)
var.set_info("POS", var.pos)
unique_name = var.var_id
# The EVENT ID is not unique. Stick with ID column.
# if 'EVENT' in var.info:
# unique_name = var.info['EVENT']
if var.info['SVTYPE'] != 'BND':
bedpe_out.write(str(converter.convert(var)) + '\n')
else:
# Manta doesn't use the SECONDARY flag
# So, just check whether the mate ID is in bnds
# Which means that the mate has already been seen
if 'MATEID' in var.info:
mate_id = var.info['MATEID']
if mate_id in bnds:
var1 = bnds[mate_id]
bedpe_out.write(str(converter.convert(var1, var)) + '\n')
del bnds[mate_id]
else:
bnds.update({unique_name:var})
else:
bnds.update({unique_name:var})
continue
intersected_keys = bnds.viewkeys() & sec_bnds.viewkeys()
for key in intersected_keys:
bedpe_out.write(str(converter.convert(bnds[key], sec_bnds[key])) + '\n')
del bnds[key]
del sec_bnds[key]
if bnds is not None:
for bnd in bnds:
sys.stderr.write('Warning: missing secondary multiline variant at ID:' + bnd + '\n')
bedpe_out.write(str(converter.convert(bnds[bnd], None)) + '\n')
if sec_bnds is not None:
for bnd in sec_bnds:
sys.stderr.write('Warning: missing primary multiline variant at ID:' + bnd + '\n')
bedpe_out.write(str(converter.convert(None, sec_bnds[bnd])) + '\n')
# close the files
bedpe_out.close()
return
class TestVcfToBedpeConverter(TestCase):
def setUp(self):
self.converter = VcfToBedpeConverter()
header_lines = [
'##fileformat=VCFv4.2', '##fileDate=20090805',
'##reference=file:///seq/references/1000GenomesPilot-NCBI36.fasta',
'##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">',
'##INFO=<ID=AF,Number=A,Type=Float,Description="Allele Frequency">',
'##INFO=<ID=END,Number=1,Type=Integer,Description="End position of the variant">',
'##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">',
'#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT NA00001'
]
self.vcf = Vcf()
self.vcf.add_header(header_lines)
def test_bnd_alt_string(self):
self.assertEqual(self.converter.parse_bnd_alt_string('A[1:6['),
('[', '1', 6))
self.assertEqual(self.converter.parse_bnd_alt_string('A]1:6]'),
(']', '1', 6))
self.assertEqual(self.converter.parse_bnd_alt_string(']1:6]A'),
(']', '1', 6))
with self.assertRaises(AssertionError):
self.converter.parse_bnd_alt_string(']1:6[A')
with self.assertRaises(AssertionError):
self.converter.parse_bnd_alt_string('1')
def test_bnd_breakpoints(self):
vcf_array1 = [
'1', '20000', '235', 'T', 'A[1:6[', '0.00', '.', '.', 'GT', '0/0'
]
v1 = Variant(vcf_array1, self.vcf)
self.assertEqual(self.converter.bnd_breakpoints(v1),
('1', 20000, 20000, '1', 5, 5, '+', '-'))
vcf_array2 = [
'1', '20000', '235', 'T', ']1:6]N', '0.00', '.', '.', 'GT', '0/0'
]
v2 = Variant(vcf_array2, self.vcf)
self.assertEqual(self.converter.bnd_breakpoints(v2),
('1', 19999, 19999, '1', 6, 6, '-', '+'))
def test_simple_breakpoints(self):
vcf_array1 = [
'1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'END=20500', 'GT',
'0/0'
]
v1 = Variant(vcf_array1, self.vcf)
self.assertEqual(self.converter.simple_breakpoints(v1),
('1', 20000, 20000, '1', 20500, 20500, '+', '-'))
vcf_array2 = [
'1', '20000', '235', 'T', '<DEL>', '0.00', '.',
'END=20500;STRANDS=-+:2', 'GT', '0/0'
]
v2 = Variant(vcf_array2, self.vcf)
self.assertEqual(self.converter.simple_breakpoints(v2),
('1', 20000, 20000, '1', 20500, 20500, '-', '+'))
vcf_array3 = [
'1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'STRANDS=--:2',
'GT', '0/0'
]
v3 = Variant(vcf_array3, self.vcf)
with self.assertRaises(ValueError):
self.converter.simple_breakpoints(v3)
def test_adjust_coordinate(self):
vcf_array1 = [
'1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'CIEND=-50,50',
'GT', '0/0'
]
v1 = Variant(vcf_array1, self.vcf)
self.assertEqual(
self.converter.adjust_coordinate(v1, 'CIEND', 500, 1000),
(450, 1050))
self.assertEqual(
self.converter.adjust_coordinate(v1, 'CIPOS', 500, 1000),
(500, 1000))
vcf_array2 = [
'1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'CIEND=50', 'GT',
'0/0'
]
v2 = Variant(vcf_array2, self.vcf)
with self.assertRaises(ValueError):
self.converter.adjust_coordinate(v2, 'CIEND', 500, 1000)
class TestVcfToBedpeConverter(TestCase):
def setUp(self):
self.converter = VcfToBedpeConverter()
header_lines = [
'##fileformat=VCFv4.2',
'##fileDate=20090805',
'##reference=file:///seq/references/1000GenomesPilot-NCBI36.fasta',
'##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">',
'##INFO=<ID=AF,Number=A,Type=Float,Description="Allele Frequency">',
'##INFO=<ID=END,Number=1,Type=Integer,Description="End position of the variant">',
'##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">',
'#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT NA00001']
self.vcf = Vcf()
self.vcf.add_header(header_lines)
def test_bnd_alt_string(self):
self.assertEqual(self.converter.parse_bnd_alt_string('A[1:6['), ('[', '1', 6))
self.assertEqual(self.converter.parse_bnd_alt_string('A]1:6]'), (']', '1', 6))
self.assertEqual(self.converter.parse_bnd_alt_string(']1:6]A'), (']', '1', 6))
with self.assertRaises(AssertionError):
self.converter.parse_bnd_alt_string(']1:6[A')
with self.assertRaises(AssertionError):
self.converter.parse_bnd_alt_string('1')
def test_bnd_breakpoints(self):
vcf_array1 = ['1', '20000', '235', 'T', 'A[1:6[', '0.00', '.', '.', 'GT', '0/0']
v1 = Variant(vcf_array1, self.vcf)
self.assertEqual(
self.converter.bnd_breakpoints(v1),
('1', 20000, 20000, '1', 5, 5, '+', '-'))
vcf_array2 = ['1', '20000', '235', 'T', ']1:6]N', '0.00', '.', '.', 'GT', '0/0']
v2 = Variant(vcf_array2, self.vcf)
self.assertEqual(
self.converter.bnd_breakpoints(v2),
('1', 19999, 19999, '1', 6, 6, '-', '+'))
def test_simple_breakpoints(self):
vcf_array1 = ['1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'END=20500', 'GT', '0/0']
v1 = Variant(vcf_array1, self.vcf)
self.assertEqual(
self.converter.simple_breakpoints(v1),
('1', 20000, 20000, '1', 20500, 20500, '+', '-'))
vcf_array2 = ['1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'END=20500;STRANDS=-+:2', 'GT', '0/0']
v2 = Variant(vcf_array2, self.vcf)
self.assertEqual(
self.converter.simple_breakpoints(v2),
('1', 20000, 20000, '1', 20500, 20500, '-', '+'))
vcf_array3 = ['1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'STRANDS=--:2', 'GT', '0/0']
v3 = Variant(vcf_array3, self.vcf)
with self.assertRaises(ValueError):
self.converter.simple_breakpoints(v3)
def test_adjust_coordinate(self):
vcf_array1 = ['1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'CIEND=-50,50', 'GT', '0/0']
v1 = Variant(vcf_array1, self.vcf)
self.assertEqual(
self.converter.adjust_coordinate(v1, 'CIEND', 500, 1000),
(450, 1050))
self.assertEqual(
self.converter.adjust_coordinate(v1, 'CIPOS', 500, 1000),
(500, 1000))
vcf_array2 = ['1', '20000', '235', 'T', '<DEL>', '0.00', '.', 'CIEND=50', 'GT', '0/0']
v2 = Variant(vcf_array2, self.vcf)
with self.assertRaises(ValueError):
self.converter.adjust_coordinate(v2, 'CIEND', 500, 1000)
def vcfToBedpe(vcf_file, bedpe_out):
converter = VcfToBedpeConverter()
vcf = svtools.vcf.file.Vcf()
in_header = True
header = []
sample_list = []
bnds = dict()
sec_bnds = dict()
v = []
for line in vcf_file:
if in_header:
if line[0:2] == '##':
if line.split('=')[0] == '##fileformat':
line = '##fileformat=' + "BEDPE" + '\n'
if line.split('=')[0] == '##fileDate':
line = '##fileDate=' + time.strftime('%Y%m%d') + '\n'
header.append(line)
continue
elif line[0] == '#' and line[1] != '#':
sample_list = line.rstrip().split('\t')[9:]
header.append(line)
continue
else:
# print header
in_header = False
vcf.add_header(header)
if "SVTYPE" in [info.id for info in vcf.info_list]:
vcf.add_info_after("SVTYPE", "POS", 1, 'Integer', 'Position of the variant described in this record')
header=vcf.get_header()
bedpe_out.write(header[:header.rfind('\n')] + '\n')
final_header_line = ['#CHROM_A',
'START_A',
'END_A',
'CHROM_B',
'START_B',
'END_B',
'ID',
'QUAL',
'STRAND_A',
'STRAND_B',
'TYPE',
'FILTER',
'NAME_A',
'REF_A',
'ALT_A',
'NAME_B',
'REF_B',
'ALT_B',
'INFO_A',
'INFO_B']
if len(sample_list) > 0:
bedpe_out.write('\t'.join(final_header_line + ['FORMAT','\t'.join(map(str,sample_list))]) + '\n')
else:
bedpe_out.write('\t'.join(final_header_line) + '\n')
v = line.rstrip().split('\t')
var = svtools.vcf.variant.Variant(v, vcf)
var.set_info("POS", var.pos)
# If there is no MATEID then assume this is a single-ended BND and simply output
if var.info['SVTYPE'] != 'BND' or 'MATEID' not in var.info:
bedpe_out.write(str(converter.convert(var)) + '\n')
else:
mate_id = var.info['MATEID']
if 'SECONDARY' in var.info:
if mate_id in bnds:
#primary
var1 = bnds[mate_id]
bedpe_out.write(str(converter.convert(var1, var)) + '\n')
del bnds[mate_id]
else:
sec_bnds.update({var.var_id:var})
else:
if mate_id in sec_bnds:
var2 = sec_bnds[mate_id]
bedpe_out.write(str(converter.convert(var, var2)) + '\n')
del sec_bnds[mate_id]
else:
bnds.update({var.var_id:var})
if bnds is not None:
for bnd in bnds:
sys.stderr.write('Warning: missing secondary multiline variant at ID:' + bnd + '\n')
bedpe_out.write(str(converter.convert(bnds[bnd], None)) + '\n')
if sec_bnds is not None:
for bnd in sec_bnds:
sys.stderr.write('Warning: missing primary multiline variant at ID:' + bnd + '\n')
bedpe_out.write(str(converter.convert(None, sec_bnds[bnd])) + '\n')
# close the files
bedpe_out.close()
return
def vcfToBedpe(vcf_file, bedpe_out):
converter = VcfToBedpeConverter()
vcf = svtools.vcf.file.Vcf()
in_header = True
header = []
sample_list = []
bnds = dict()
sec_bnds = dict()
v = []
for line in vcf_file:
if in_header:
if line[0:2] == '##':
if line.split('=')[0] == '##fileformat':
line = '##fileformat=' + "BEDPE" + '\n'
if line.split('=')[0] == '##fileDate':
line = '##fileDate=' + time.strftime('%Y%m%d') + '\n'
header.append(line)
continue
elif line[0] == '#' and line[1] != '#':
sample_list = line.rstrip().split('\t')[9:]
header.append(line)
continue
else:
# print header
in_header = False
vcf.add_header(header)
if "SVTYPE" in [info.id for info in vcf.info_list]:
vcf.add_info_after(
"SVTYPE", "POS", 1, 'Integer',
'Position of the variant described in this record')
header = vcf.get_header()
bedpe_out.write(header[:header.rfind('\n')] + '\n')
final_header_line = [
'#CHROM_A', 'START_A', 'END_A', 'CHROM_B', 'START_B',
'END_B', 'ID', 'QUAL', 'STRAND_A', 'STRAND_B', 'TYPE',
'FILTER', 'NAME_A', 'REF_A', 'ALT_A', 'NAME_B', 'REF_B',
'ALT_B', 'INFO_A', 'INFO_B'
]
if len(sample_list) > 0:
bedpe_out.write('\t'.join(
final_header_line +
['FORMAT', '\t'.join(map(str, sample_list))]) + '\n')
else:
bedpe_out.write('\t'.join(final_header_line) + '\n')
v = line.rstrip().split('\t')
var = svtools.vcf.variant.Variant(v, vcf)
var.set_info("POS", var.pos)
# If there is no MATEID then assume this is a single-ended BND and simply output
if var.info['SVTYPE'] != 'BND' or 'MATEID' not in var.info:
bedpe_out.write(str(converter.convert(var)) + '\n')
else:
mate_id = var.info['MATEID']
if 'SECONDARY' in var.info:
if mate_id in bnds:
#primary
var1 = bnds[mate_id]
bedpe_out.write(str(converter.convert(var1, var)) + '\n')
del bnds[mate_id]
else:
sec_bnds.update({var.var_id: var})
else:
if mate_id in sec_bnds:
var2 = sec_bnds[mate_id]
bedpe_out.write(str(converter.convert(var, var2)) + '\n')
del sec_bnds[mate_id]
elif mate_id in bnds:
var1 = bnds[mate_id]
bedpe_out.write(str(converter.convert(var1, var)) + '\n')
del bnds[mate_id]
else:
bnds.update({var.var_id: var})
if bnds is not None:
for bnd in bnds:
sys.stderr.write(
'Warning: missing secondary multiline variant at ID:' + bnd +
'\n')
bedpe_out.write(str(converter.convert(bnds[bnd], None)) + '\n')
if sec_bnds is not None:
for bnd in sec_bnds:
sys.stderr.write(
'Warning: missing primary multiline variant at ID:' + bnd +
'\n')
bedpe_out.write(str(converter.convert(None, sec_bnds[bnd])) + '\n')
# close the files
bedpe_out.close()
return
def vcfToBedpe(vcf_file, bedpe_out):
converter = VcfToBedpeConverter()
vcf = svtools.vcf.file.Vcf()
in_header = True
header = []
sample_list = []
bnds = dict()
sec_bnds = dict()
v = []
for line in vcf_file:
if in_header:
if line[0:2] == '##':
if line.split('=')[0] == '##fileformat':
line = '##fileformat=' + "BEDPE" + '\n'
if line.split('=')[0] == '##fileDate':
line = '##fileDate=' + time.strftime('%Y%m%d') + '\n'
header.append(line)
continue
elif line[0] == '#' and line[1] != '#':
sample_list = line.rstrip().split('\t')[9:]
header.append(line)
continue
else:
# print header
in_header = False
vcf.add_header(header)
if "SVTYPE" in [info.id for info in vcf.info_list]:
vcf.add_info_after(
"SVTYPE", "POS", 1, 'Integer',
'Position of the variant described in this record')
header = vcf.get_header()
bedpe_out.write(header[:header.rfind('\n')] + '\n')
final_header_line = [
'#CHROM_A', 'START_A', 'END_A', 'CHROM_B', 'START_B',
'END_B', 'ID', 'QUAL', 'STRAND_A', 'STRAND_B', 'TYPE',
'FILTER', 'NAME_A', 'REF_A', 'ALT_A', 'NAME_B', 'REF_B',
'ALT_B', 'INFO_A', 'INFO_B'
]
if len(sample_list) > 0:
bedpe_out.write('\t'.join(
final_header_line +
['FORMAT', '\t'.join(map(str, sample_list))]) + '\n')
else:
bedpe_out.write('\t'.join(final_header_line) + '\n')
v = line.rstrip().split('\t')
var = svtools.vcf.variant.Variant(v, vcf)
var.set_info("POS", var.pos)
unique_name = var.var_id
# The EVENT ID is not unique. Stick with ID column.
# if 'EVENT' in var.info:
# unique_name = var.info['EVENT']
if var.info['SVTYPE'] != 'BND':
bedpe_out.write(str(converter.convert(var)) + '\n')
else:
# Manta doesn't use the SECONDARY flag
# So, just check whether the mate ID is in bnds
# Which means that the mate has already been seen
if 'MATEID' in var.info:
mate_id = var.info['MATEID']
if mate_id in bnds:
var1 = bnds[mate_id]
bedpe_out.write(str(converter.convert(var1, var)) + '\n')
del bnds[mate_id]
else:
bnds.update({unique_name: var})
else:
bnds.update({unique_name: var})
continue
intersected_keys = bnds.viewkeys() & sec_bnds.viewkeys()
for key in intersected_keys:
bedpe_out.write(
str(converter.convert(bnds[key], sec_bnds[key])) + '\n')
del bnds[key]
del sec_bnds[key]
if bnds is not None:
for bnd in bnds:
sys.stderr.write(
'Warning: missing secondary multiline variant at ID:' + bnd +
'\n')
bedpe_out.write(str(converter.convert(bnds[bnd], None)) + '\n')
if sec_bnds is not None:
for bnd in sec_bnds:
sys.stderr.write(
'Warning: missing primary multiline variant at ID:' + bnd +
'\n')
bedpe_out.write(str(converter.convert(None, sec_bnds[bnd])) + '\n')
# close the files
bedpe_out.close()
return
