def test_coordinates_for_insertion_and_deletion(self):
record = model._Record(
'1',
10,
'id10',
'CT',
[
model._Substitution('CA'),
model._Substitution('CTT'),
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 11), (10, 11))
record = model._Record(
'1',
10,
'id11',
'CT',
[
model._Substitution('CTT'),
model._Substitution('CA'),
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 11), (10, 11))
def test_coordinates_for_insert_and_snp(self):
record = model._Record(
'1',
10,
'id6',
'C',
[
model._Substitution('GTA'),
model._Substitution('G'),
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 10), (9, 10))
record = model._Record(
'1',
10,
'id7',
'C',
[
model._Substitution('G'),
model._Substitution('GTA'),
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 10), (9, 10))
def test_coordinates_for_snp_and_deletion(self):
record = model._Record(
'1',
10,
'id8',
'CTA',
[
model._Substitution('C'),
model._Substitution('CTG'),
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 12), (10, 12))
record = model._Record(
'1',
10,
'id9',
'CTA',
[
model._Substitution('CTG'),
model._Substitution('C'),
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 12), (10, 12))
def test_coordinates_for_multiple_snps(self):
record = model._Record('1', 10, 'id5', 'C', [
model._Substitution('A'),
model._Substitution('G'),
model._Substitution('T')
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 10), (9, 10))
def test_coordinates_for_insertion_and_deletion(self):
record = model._Record('1', 10, 'id10', 'CT', [
model._Substitution('CA'),
model._Substitution('CTT'),
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 11), (10, 11))
record = model._Record('1', 10, 'id11', 'CT', [
model._Substitution('CTT'),
model._Substitution('CA'),
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 11), (10, 11))
def test_coordinates_for_snp_and_deletion(self):
record = model._Record('1', 10, 'id8', 'CTA', [
model._Substitution('C'),
model._Substitution('CTG'),
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 12), (10, 12))
record = model._Record('1', 10, 'id9', 'CTA', [
model._Substitution('CTG'),
model._Substitution('C'),
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 12), (10, 12))
def test_coordinates_for_insert_and_snp(self):
record = model._Record('1', 10, 'id6', 'C', [
model._Substitution('GTA'),
model._Substitution('G'),
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 10), (9, 10))
record = model._Record('1', 10, 'id7', 'C', [
model._Substitution('G'),
model._Substitution('GTA'),
], None, None, {}, None, {}, None)
self.assert_has_expected_coordinates(record, (9, 10), (9, 10))
def _parse_alt(self, str):
if self._alt_pattern.search(str) is not None:
# Paired breakend
items = self._alt_pattern.split(str)
remoteCoords = items[1].split(':')
chr = remoteCoords[0]
if chr[0] == '<':
chr = chr[1:-1]
withinMainAssembly = False
else:
withinMainAssembly = True
pos = remoteCoords[1]
orientation = (str[0] == '[' or str[0] == ']')
remoteOrientation = (re.search('\[', str) is not None)
if orientation:
connectingSequence = items[2]
else:
connectingSequence = items[0]
return _Breakend(chr, pos, orientation, remoteOrientation, connectingSequence, withinMainAssembly)
elif str[0] == '.' and len(str) > 1:
return _SingleBreakend(True, str[1:])
elif str[-1] == '.' and len(str) > 1:
return _SingleBreakend(False, str[:-1])
elif str[0] == "<" and str[-1] == ">":
return _SV(str[1:-1])
else:
return _Substitution(str)
def _get_record_for_indel(self, bpm_record_group):
"""
Create a new VCF record for an indel
Args:
bpm_record_group (list(BPMRecord)) : BPM records for the group (must all be indels)
Returns:
vcf._Record : The new VCF record definition
"""
(qual, filt, info,
sample_indexes) = VcfRecordFactory._get_record_defaults()
identifier = self._get_identifier(bpm_record_group)
for record in bpm_record_group:
assert record.is_deletion == bpm_record_group[0].is_deletion
bpm_record = bpm_record_group[0]
(_, indel_sequence,
_) = bpm_record.get_indel_source_sequences(RefStrand.Plus)
start_index = bpm_record.pos - 1
chrom = bpm_record.chromosome
if chrom == "XX" or chrom == "XY":
chrom = "X"
if bpm_record.is_deletion:
reference_base = self._genome_reader.get_reference_bases(
chrom, start_index - 1, start_index)
reference_allele = reference_base + indel_sequence
alternate_allele = reference_base
return _Record(chrom, start_index, identifier, reference_allele,
[_Substitution(alternate_allele)], qual, filt, info,
self._format_factory.get_format_id_string(),
sample_indexes)
reference_base = self._genome_reader.get_reference_bases(
chrom, start_index, start_index + 1)
reference_allele = reference_base
alternate_allele = reference_base + indel_sequence
return _Record(chrom, start_index + 1, identifier, reference_allele,
[_Substitution(alternate_allele)], qual, filt, info,
self._format_factory.get_format_id_string(),
sample_indexes)
def _grid_item_to_vcf_record(info_dict, obj, sample_ids, sample_names): # , get_genotype_from_expanded_zygosity):
CHROM = obj.get("locus__contig__name", ".")
POS = obj.get("locus__position", ".")
ID = obj.get("variantannotation__dbsnp_rs_id")
REF = obj.get("locus__ref__seq", ".")
ALT = obj.get("alt__seq", ".")
QUAL = '.' # QUAL = obj.get("annotation__quality", ".")
FILTER = None
INFO = {}
for info_id, data in info_dict.items():
col = data['column__variant_column']
val = obj.get(col)
if val:
INFO[info_id] = val
FORMAT = None
MY_FORMAT = ['GT', 'AD', 'AF', 'PL', 'DP', 'GQ']
CallData = make_calldata_tuple(MY_FORMAT)
sample_indexes = {}
samples = []
if sample_ids:
FORMAT = ':'.join(MY_FORMAT)
alts = [_Substitution(ALT)]
ALT = alts
record = _Record(CHROM, POS, ID, REF, ALT, QUAL, FILTER, INFO, FORMAT, sample_indexes)
if sample_ids:
for i, (sample_id, sample) in enumerate(zip(sample_ids, sample_names)):
ad = obj[f"{sample_id}_samples_allele_depth"]
zygosity = obj[f"{sample_id}_samples_zygosity"]
gt = Zygosity.get_genotype_from_expanded_zygosity(zygosity)
dp = obj[f"{sample_id}_samples_read_depth"]
af = obj[f"{sample_id}_samples_allele_frequency"]
# GQ/PL/FT are optional now
# TODO: Ideally, we'd not write them out
pl = obj.get(f"{sample_id}_samples_phred_likelihood", ".")
gq = obj.get(f"{sample_id}_samples_genotype_quality", ".")
# TODO: Need to grab information for reference base to be able to properly fill in this data.
data_args = {'AD': ['.', ad],
'GT': gt,
'PL': ['.', pl],
'DP': ['.', dp],
'GQ': ['.', gq],
'AF': ['.', af]}
data = CallData(**data_args)
call = _Call(record, sample, data)
samples.append(call)
sample_indexes[sample] = i
record.samples = samples
return record
def test_coordinates_for_multiple_snps(self):
record = model._Record(
'1',
10,
'id5',
'C',
[
model._Substitution('A'),
model._Substitution('G'),
model._Substitution('T')
],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 10), (9, 10))
def _get_record_for_snv(self, bpm_record_group):
"""
Create a new VCF record for an SNV
Args:
bpm_record_group (list(BPMRecord)) : BPM records for the group (must all be SNV records)
Returns:
vcf._Record : The new VCF record definition
"""
assert not bpm_record_group[0].is_indel()
identifier = self._get_identifier(bpm_record_group)
bpm_record = bpm_record_group[0]
(qual, filt, info,
sample_indexes) = VcfRecordFactory._get_record_defaults()
start_index = bpm_record.pos - 1
chrom = bpm_record.chromosome
if chrom == "XX" or chrom == "XY":
chrom = "X"
reference_base = self._genome_reader.get_reference_bases(
chrom, start_index, start_index + 1)
if not check_reference_allele(reference_base, bpm_record_group):
self._logger.warn("Reference allele is not queried for locus: " +
identifier)
alts = []
for record in bpm_record_group:
for nucleotide in record.plus_strand_alleles:
if nucleotide != reference_base:
substitution = _Substitution(nucleotide)
if substitution not in alts:
alts.append(_Substitution(nucleotide))
return _Record(chrom, bpm_record.pos, identifier, reference_base, alts,
qual, filt, info,
self._format_factory.get_format_id_string(),
sample_indexes)
def test_coordinates_for_insertion(self):
record = model._Record(
'1',
10,
'id2',
'C',
[model._Substitution('CTA')],
None,
None,
{},
None,
{},
None
)
self.assert_has_expected_coordinates(record, (9, 10), (10, 10))
def test_is_snp_for_n_alt(self):
record = model._Record(
'1',
10,
'id1',
'C',
[model._Substitution('N')],
None,
None,
{},
None,
{},
None
)
self.assertTrue(record.is_snp)
def test_coordinates_for_insertion(self):
record = model._Record('1', 10, 'id2', 'C',
[model._Substitution('CTA')], None, None, {},
None, {}, None)
self.assert_has_expected_coordinates(record, (9, 10), (10, 10))
def test_is_snp_for_n_alt(self):
record = model._Record('1', 10, 'id1', 'C', [model._Substitution('N')],
None, None, {}, None, {}, None)
self.assertTrue(record.is_snp)
