def build_transcript(gene,
exons,
cds_start,
cds_end,
domains,
strand=None,
is_best_transcript=False,
name=None):
pre_transcript = PreTranscript(
exons,
gene=gene,
strand=strand if strand is not None else gene.get_strand(),
is_best_transcript=is_best_transcript,
name=name)
if gene is not None:
gene.unspliced_transcripts.append(pre_transcript)
for spl in pre_transcript.generate_splicing_patterns():
t = Transcript(pre_transcript, spl)
pre_transcript.spliced_transcripts.append(t)
tx = Translation(cds_start, cds_end, t, domains=domains)
t.translations.append(tx)
return pre_transcript
def distance_setup():
n = argparse.Namespace()
n.transcript = PreTranscript(
[(1001, 1100), (1501, 1600), (2001, 2100), (2201, 2300)], strand='+'
)
for patt in n.transcript.generate_splicing_patterns():
n.transcript.transcripts.append(Transcript(n.transcript, patt))
n.trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11,
overlapping_transcripts={n.transcript},
)
setattr(
n.trans_evidence,
'_select_transcripts',
lambda *pos: n.trans_evidence.overlapping_transcripts,
)
setattr(
n.trans_evidence,
'distance',
partial(TranscriptomeEvidence.distance, n.trans_evidence),
)
return n
def test_shift_overaligned(self):
# qwertyuiopas---kkkkk------dfghjklzxcvbnm
# .......... ................
gene = Gene('1', 1, 1000, strand='+')
transcript = PreTranscript(exons=[(1, 12), (20, 28)], gene=gene, strand='+')
for spl_patt in transcript.generate_splicing_patterns():
transcript.transcripts.append(Transcript(transcript, spl_patt))
gene.transcripts.append(transcript)
read = SamRead(
reference_name='1',
reference_start=0,
cigar=_cigar.convert_string_to_cigar('14=7D12='),
query_sequence='qwertyuiopasdfghjklzxcvbnm',
)
evidence = TranscriptomeEvidence(
annotations={},
reference_genome={'1': MockObject(seq='qwertyuiopasdfkkkkkdfghjklzxcvbnm')},
bam_cache=MockObject(get_read_reference_name=lambda r: r.reference_name),
break1=Breakpoint('1', 1, orient='L', strand='+'),
break2=Breakpoint('1', 10, orient='R', strand='+'),
read_length=75,
stdev_fragment_size=75,
median_fragment_size=220,
)
evidence.overlapping_transcripts.add(transcript)
new_read = evidence.standardize_read(read)
assert new_read.cigar == _cigar.convert_string_to_cigar('12=7N14=')
def test_net_zero(self):
transcript = PreTranscript([(1001, 1100), (1301, 1400), (1701, 1800)], strand=STRAND.POS)
for patt in transcript.generate_splicing_patterns():
transcript.transcripts.append(Transcript(transcript, patt))
trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11,
overlapping_transcripts={transcript},
)
setattr(
trans_evidence,
'_select_transcripts',
lambda *pos: trans_evidence.overlapping_transcripts,
)
setattr(
trans_evidence,
'distance',
partial(TranscriptomeEvidence.distance, trans_evidence),
)
bpp = BreakpointPair(
Breakpoint('1', 1099, orient=ORIENT.LEFT),
Breakpoint('1', 1302, orient=ORIENT.RIGHT),
untemplated_seq='TT',
)
dist = partial(TranscriptomeEvidence.distance, trans_evidence)
assert bpp.net_size() == Interval(-200)
assert bpp.net_size(dist) == Interval(0)
def trans_window_setup():
n = argparse.Namespace()
gene = Gene('1', 1, 9999, name='KRAS', strand=STRAND.POS)
n.pre_transcript = PreTranscript(
gene=gene, exons=[(1001, 1100), (1401, 1500), (1701, 1750), (3001, 4000)]
)
gene.unspliced_transcripts.append(n.pre_transcript)
for spl in n.pre_transcript.generate_splicing_patterns():
n.pre_transcript.transcripts.append(Transcript(n.pre_transcript, spl))
n.annotations = {gene.chr: [gene]}
n.genome_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
config={**DEFAULTS, 'validate.call_error': 11},
)
n.trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
overlapping_transcripts={n.pre_transcript},
config={**DEFAULTS, 'validate.call_error': 11},
)
setattr(
n.trans_evidence,
'_select_transcripts',
lambda *pos: n.trans_evidence.overlapping_transcripts,
)
setattr(
n.trans_evidence,
'traverse',
partial(TranscriptomeEvidence.traverse, n.trans_evidence),
)
return n
def setUp(self):
gene = Gene('1', 1, 9999, name='KRAS', strand=STRAND.POS)
self.pre_transcript = PreTranscript(gene=gene,
exons=[(1001, 1100), (1401, 1500),
(1701, 1750), (3001, 4000)])
gene.unspliced_transcripts.append(self.pre_transcript)
for spl in self.pre_transcript.generate_splicing_patterns():
self.pre_transcript.transcripts.append(
Transcript(self.pre_transcript, spl))
self.annotations = {gene.chr: [gene]}
self.genome_evidence = MockObject(annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11)
self.trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11,
overlapping_transcripts={self.pre_transcript},
)
setattr(
self.trans_evidence,
'_select_transcripts',
lambda *pos: self.trans_evidence.overlapping_transcripts,
)
setattr(
self.trans_evidence,
'traverse',
partial(TranscriptomeEvidence.traverse, self.trans_evidence),
)
def setUp(self):
self.transcript = PreTranscript([(1001, 1100), (1301, 1400),
(1701, 1800)],
strand=STRAND.NEG)
for patt in self.transcript.generate_splicing_patterns():
self.transcript.transcripts.append(
Transcript(self.transcript, patt))
self.trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11,
overlapping_transcripts={self.transcript},
)
setattr(
self.trans_evidence,
'_select_transcripts',
lambda *pos: self.trans_evidence.overlapping_transcripts,
)
setattr(
self.trans_evidence,
'traverse',
partial(TranscriptomeEvidence.traverse, self.trans_evidence),
)
def test_empty_intron(self, distance_setup):
t2 = PreTranscript([(1001, 1100), (1501, 1600), (2001, 2200), (2201, 2300)], strand='+')
for patt in t2.generate_splicing_patterns():
t2.transcripts.append(Transcript(t2, patt))
print(t2)
print(distance_setup.trans_evidence.overlapping_transcripts)
distance_setup.trans_evidence.overlapping_transcripts.add(t2)
dist = distance_setup.trans_evidence.distance(1001, 2301)
assert dist == Interval(400, 400)
def test_multiple_transcripts(self, trans_window_setup):
# [(1001, 1100), (1401, 1500), (1701, 1750), (3001, 4000)])
b = Breakpoint(chr='1', start=1150, orient=ORIENT.RIGHT)
gene = trans_window_setup.annotations['1'][0]
t2 = PreTranscript(gene=gene, exons=[(1001, 1100), (1200, 1300), (2100, 2200)])
for patt in t2.generate_splicing_patterns():
t2.transcripts.append(Transcript(t2, patt))
gene.transcripts.append(t2)
# 989 - 2561
# 989 - 3411
assert transcriptome_window(
trans_window_setup.trans_evidence, b, [trans_window_setup.pre_transcript, t2]
) == Interval(1040, 3160)
def setUp(self):
self.transcript = PreTranscript([(1001, 1100), (1501, 1600),
(2001, 2100), (2201, 2300)],
strand='+')
for patt in self.transcript.generate_splicing_patterns():
self.transcript.transcripts.append(
Transcript(self.transcript, patt))
self.trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11,
overlapping_transcripts={self.transcript})
setattr(self.trans_evidence, '_select_transcripts',
lambda *pos: self.trans_evidence.overlapping_transcripts)
setattr(self.trans_evidence, 'distance',
partial(TranscriptomeEvidence.distance, self.trans_evidence))
def test_many_small_exons(self):
g = Gene('fake', 17271277, 17279592, strand='+')
pre_transcript = PreTranscript(
gene=g,
exons=[
(17271277, 17271984),
(17272649, 17272709),
(17275586, 17275681),
(17275769, 17275930),
(17276692, 17276817),
(17277168, 17277388), # 220
(17277845, 17277888), # 44
(17278293, 17278378), # 86
(17279229, 17279592) # 364
])
g.transcripts.append(pre_transcript)
for patt in pre_transcript.generate_splicing_patterns():
pre_transcript.transcripts.append(Transcript(pre_transcript, patt))
b = Breakpoint(chr='fake', start=17279591, orient=ORIENT.LEFT)
self.assertEqual(Interval(17277321, 17279701),
self.transcriptome_window(b, [pre_transcript]))
def tranverse_trans_rev_setup():
n = argparse.Namespace()
n.transcript = PreTranscript([(1001, 1100), (1301, 1400), (1701, 1800)], strand=STRAND.NEG)
for patt in n.transcript.generate_splicing_patterns():
n.transcript.transcripts.append(Transcript(n.transcript, patt))
n.trans_evidence = MockObject(
annotations={},
read_length=100,
max_expected_fragment_size=550,
call_error=11,
overlapping_transcripts={n.transcript},
)
setattr(
n.trans_evidence,
'_select_transcripts',
lambda *pos: n.trans_evidence.overlapping_transcripts,
)
setattr(
n.trans_evidence,
'traverse',
partial(TranscriptomeEvidence.traverse, n.trans_evidence),
)
return n