def test_init(self):
bp = Breakpoint(self.entry, fixed_slop=1)
self.assertEqual(bp.l, self.entry)
self.assertEqual(bp.sv_type, 'BND')
self.assertEqual(bp.left.chrom, '1')
self.assertEqual(bp.right.chrom, '10')
self.assertEqual(bp.strands, '++:5')
self.assertEqual(bp.left.start, 9572383 - 11)
self.assertEqual(bp.left.end, 9572383 + 11)
self.assertEqual(bp.right.start, 94079366 - 11)
self.assertEqual(bp.right.end, 94079366 + 11)
self.assertEqual(bp.left.p, self.prpos)
self.assertEqual(bp.right.p, self.prend)
# This was previously implemented in l_bp_tests, adding in here too
test_line = '1 1000 2345_1 N [2:1100[N 0.00 . SVTYPE=BND;STRANDS=--:7;IMPRECISE;CIPOS=-2,2;CIEND=-2,2;CIPOS95=-1,1;CIEND95=-1,1;MATEID=2345_2;EVENT=2345;SU=7;PE=7;SR=0;PRPOS=0.025,0.25,0.45,0.25,0.025;PREND=0.025,0.25,0.45,0.25,0.025'
no_slop = Breakpoint(test_line)
self.assertEqual(no_slop.left.p, [0.025, 0.25, 0.45, 0.25, 0.025])
self.assertEqual(no_slop.right.p, [0.025, 0.25, 0.45, 0.25, 0.025])
fixed_slop = Breakpoint(test_line, fixed_slop = 1)
self.assertEqual(fixed_slop.left.p, [1e-100, 0.025, 0.25, 0.45, 0.25, 0.025, 1e-100])
self.assertEqual(fixed_slop.right.p, [1e-100, 0.025, 0.25, 0.45, 0.25, 0.025, 1e-100])
percent_slop = Breakpoint(test_line, percent_slop = 0.2)
print percent_slop
self.assertEqual(percent_slop.left.p, [1e-100, 0.025, 0.25, 0.45, 0.25, 0.025, 1e-100])
self.assertEqual(percent_slop.right.p, [1e-100, 0.025, 0.25, 0.45, 0.25, 0.025, 1e-100])
percent_and_fixed_slop = Breakpoint(test_line, percent_slop = 0.2, fixed_slop = 2)
self.assertEqual(percent_and_fixed_slop.left.p, [1e-100, 1e-100, 0.025, 0.25, 0.45, 0.25, 0.025, 1e-100, 1e-100])
self.assertEqual(percent_and_fixed_slop.right.p, [1e-100, 1e-100, 0.025, 0.25, 0.45, 0.25, 0.025, 1e-100, 1e-100])
def test_str(self):
bp = Breakpoint(self.entry, fixed_slop=1)
expected = [
'1',
str(9572383 - 11),
str(9572383 + 11), '10',
str(94079366 - 11),
str(94079366 + 11), 'BND', '++:5',
str(self.prpos),
str(self.prend)
]
self.assertEqual(str(bp), '\t'.join(expected))
def l_cluster_by_line(file_name,
tempdir,
percent_slop=0,
fixed_slop=0,
use_product=False,
include_genotypes=False,
weighting_scheme='unweighted'):
v_id = 0
in_header = True
header = []
vcf = Vcf()
vcf_out = sys.stdout
with InputStream(file_name, tempdir) as vcf_stream:
BP_l = []
BP_sv_type = ''
BP_max_end_l = -1
BP_chr_l = ''
sample_order = []
for line in vcf_stream:
if in_header:
if line.startswith('##'):
header.append(line)
continue
elif line.startswith('#CHROM'):
v = line.rstrip().split('\t')
for headline in header:
if headline[:8] == '##SAMPLE':
sample_order.append(headline.rstrip()[13:-1])
hline = ''
if include_genotypes:
v.extend(sample_order)
hline = '\t'.join(v)
else:
v = v[:8]
hline = '\t'.join(v)
header.append(hline)
in_header = False
vcf.add_header(header)
vcf.add_info('ALG', '1', 'String',
'Algorithm used to merge this breakpoint')
if include_genotypes:
vcf_out.write(vcf.get_header() + '\n')
else:
vcf_out.write(vcf.get_header(False) + '\n')
continue
b = Breakpoint(l_bp.parse_vcf_record(line),
percent_slop=percent_slop,
fixed_slop=fixed_slop)
if (len(BP_l) == 0) or ((b.left.start <= BP_max_end_l) and
(b.left.chrom == BP_chr_l) and
(b.sv_type == BP_sv_type)):
BP_l.append(b)
BP_max_end_l = max(BP_max_end_l, b.left.end)
BP_chr_l = b.left.chrom
BP_sv_type = b.sv_type
else:
v_id = r_cluster(BP_l, sample_order, v_id, use_product, vcf,
vcf_out, include_genotypes, weighting_scheme)
BP_l = [b]
BP_max_end_l = b.left.end
BP_sv_type = b.sv_type
BP_chr_l = b.left.chrom
if len(BP_l) > 0:
v_id = r_cluster(BP_l, sample_order, v_id, use_product, vcf,
vcf_out, include_genotypes, weighting_scheme)
def l_cluster_by_line(file_name,
percent_slop=0,
fixed_slop=0,
use_product=False):
v_id = 0
vcf_lines = []
vcf_headers = list()
infile = open(file_name, 'r')
header = ''
samples = ''
for l in infile:
if l[0] == '#':
if l[1] != '#':
samples = l.rstrip().split('\t')[9:]
else:
# ignore fileDate
if l[:10] == '##fileDate':
continue
if l not in vcf_headers:
vcf_headers.append(l)
if l[0] != '#':
break
vcf_headers.append("#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\n")
sample_order = []
for header in vcf_headers:
if header[:8] == '##SAMPLE':
sample_order.append(header.rstrip()[13:-1])
print header,
BP_l = []
BP_sv_type = ''
BP_max_end_l = -1
BP_chr_l = ''
b = Breakpoint(l_bp.parse_vcf_record(l),
percent_slop=percent_slop,
fixed_slop=fixed_slop)
BP_l.append(b)
BP_max_end_l = max(BP_max_end_l, b.end_l)
BP_chr_l = b.chr_l
BP_sv_type = b.sv_type
for l in infile:
b = Breakpoint(l_bp.parse_vcf_record(l),
percent_slop=percent_slop,
fixed_slop=fixed_slop)
if (len(BP_l) == 0) or ((b.start_l <= BP_max_end_l) and
(b.chr_l == BP_chr_l) and
(b.sv_type == BP_sv_type)):
BP_l.append(b)
BP_max_end_l = max(BP_max_end_l, b.end_l)
BP_chr_l = b.chr_l
BP_sv_type = b.sv_type
else:
v_id = r_cluster(BP_l, sample_order, v_id, use_product)
BP_l = [b]
BP_max_end_l = b.end_l
BP_sv_type = b.sv_type
BP_chr_l = b.chr_l
if len(BP_l) > 0:
v_id = r_cluster(BP_l, sample_order, v_id, use_product)
infile.close()
def test_ovl(self):
bp = Breakpoint(self.entry, fixed_slop=1)
bp2 = Breakpoint(self.entry, fixed_slop=2)
# Note that this is a regression test. This value was arrived at using the existing code.
# It's correctness is unknown.
self.assertEqual(bp.ovl(bp2), 1.0)
def l_cluster_by_line(file_name,
tempdir,
percent_slop=0,
fixed_slop=0,
use_product=False,
include_genotypes=False,
weighting_scheme='unweighted'):
v_id = 0
in_header = True
header = []
vcf = Vcf()
vcf_out = sys.stdout
with InputStream(file_name, tempdir) as vcf_stream:
BP_l = []
BP_sv_type = ''
BP_max_end_l = -1
BP_chr_l = ''
sample_order = []
for line in vcf_stream:
if in_header:
if line.startswith('##'):
header.append(line)
continue
elif line.startswith('#CHROM'):
v = line.rstrip().split(
'\t') # #CHROM line split -> list -D
for headline in header:
if headline[:8] == '##SAMPLE' and headline.rstrip(
)[13:-1] != 'VARIOUS':
sample_order.append(
headline.rstrip()[13:-1]
) # maybe add sample name to samplr_order list. -D
hline = '' # Parsed #CHROM line from 'v' -D
if include_genotypes:
v = v[:9] # Remove possible VARIOUS -D
v.extend(sample_order)
hline = '\t'.join(v)
else:
v = v[:8] # No FORMAT field here. -D
hline = '\t'.join(v)
header.append(hline)
in_header = False
vcf.add_header(header)
vcf.add_info('ALG', '1', 'String',
'Algorithm used to merge this breakpoint')
if include_genotypes:
vcf_out.write(vcf.get_header() + '\n')
else:
vcf_out.write(
vcf.get_header(include_samples=False) +
'\n') # Not including samples here. -D
continue
# Header is now parsed, then the main dish. -D
b = Breakpoint(
l_bp.parse_vcf_record(line),
percent_slop=percent_slop,
fixed_slop=fixed_slop) # percent_slop and fixed_slop is 0. -D
if (len(BP_l) == 0) or (
(b.left.start <= BP_max_end_l) and
(b.left.chrom == BP_chr_l) and (b.sv_type == BP_sv_type)
): # Same chrom svtype and start is small than previous end. -D
BP_l.append(b)
BP_max_end_l = max(BP_max_end_l, b.left.end)
BP_chr_l = b.left.chrom
BP_sv_type = b.sv_type
else:
v_id = r_cluster(BP_l, sample_order, v_id, use_product, vcf,
vcf_out, include_genotypes, weighting_scheme)
BP_l = [b]
BP_max_end_l = b.left.end
BP_chr_l = b.left.chrom
BP_sv_type = b.sv_type
if len(BP_l) > 0:
v_id = r_cluster(BP_l, sample_order, v_id, use_product, vcf,
vcf_out, include_genotypes, weighting_scheme)
def test_ovl(self):
bp = Breakpoint(self.entry, fixed_slop=1)
bp2 = Breakpoint(self.entry, fixed_slop=2)
# Note that this is a regression test. This value was arrived at using the existing code.
# It's correctness is unknown.
self.assertEqual(bp.ovl(bp2), 1.0)
def test_floats_from_tag(self):
bp = Breakpoint(self.entry, fixed_slop=1)
info = { 'TAG': '1.2,1.3'}
self.assertEqual(bp.floats_from_tag(info, 'TAG'), [1.2, 1.3])
with self.assertRaises(RuntimeError):
bp.floats_from_tag(info, 'AG')