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 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 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 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 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_by_strand(self, strand):
self.ex1 = Exon(100, 199, strand=strand) # C
self.ex2 = Exon(500, 599, strand=strand) # G
self.ex3 = Exon(1200, 1299, strand=strand) # T
self.ex4 = Exon(1500, 1599, strand=strand) # C
self.ex5 = Exon(1700, 1799, strand=strand) # G
self.ex6 = Exon(2000, 2099, strand=strand) # C
# introns: 99, 300, 600, 200, 100, ...
reference_sequence = 'a' * 99 + 'C' * 100 + 'a' * 300 + 'G' * 100
reference_sequence += 'a' * 600 + 'T' * 100 + 'a' * 200 + 'C' * 100
reference_sequence += 'a' * 100 + 'G' * 100 + 'a' * 200 + 'C' * 100
self.reference_sequence = reference_sequence
self.pre_transcript = PreTranscript(exons=[self.ex1, self.ex2, self.ex3, self.ex4, self.ex5, self.ex6], strand=strand)
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 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
class TestNetSizeTrans(unittest.TestCase):
def setUp(self):
self.transcript = PreTranscript([(1001, 1100), (1301, 1400),
(1701, 1800)],
strand=STRAND.POS)
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_net_zero(self):
bpp = BreakpointPair(Breakpoint('1', 1099, orient=ORIENT.LEFT),
Breakpoint('1', 1302, orient=ORIENT.RIGHT),
untemplated_seq='TT')
dist = partial(TranscriptomeEvidence.distance, self.trans_evidence)
self.assertEqual(Interval(-200), bpp.net_size())
self.assertEqual(Interval(0), bpp.net_size(dist))
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))
class TestDistance(unittest.TestCase):
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_exonic(self):
self.assertEqual(Interval(149),
self.trans_evidence.distance(1001, 1550))
def test_intergenic_exonic(self):
dist = self.trans_evidence.distance(101, 1550)
self.assertEqual(Interval(1049, 1049), dist)
def test_intergenic_intergenic(self):
dist = self.trans_evidence.distance(101, 300)
self.assertEqual(Interval(199), dist)
def test_aligned_intronic(self):
dist = self.trans_evidence.distance(1102, 1499)
self.assertEqual(Interval(5), dist)
def test_indel_at_exon_boundary(self):
self.assertEqual(Interval(2), self.trans_evidence.distance(1101, 1501))
def test_no_annotations(self):
dist = self.trans_evidence.distance(101, 300, [])
self.assertEqual(Interval(199), dist)
def test_intergenic_intronic(self):
dist = self.trans_evidence.distance(101, 1400)
self.assertEqual(Interval(1101), dist)
def test_empty_intron(self):
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(self.trans_evidence.overlapping_transcripts)
self.trans_evidence.overlapping_transcripts.add(t2)
dist = self.trans_evidence.distance(1001, 2301)
self.assertEqual(Interval(400, 400), dist)
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 setUp(self):
self.gev1 = BreakpointPair(
Breakpoint('1', 1),
Breakpoint('1', 10),
opposing_strands=True,
data={
COLUMNS.event_type: SVTYPE.DEL,
COLUMNS.call_method: CALL_METHOD.CONTIG,
COLUMNS.fusion_sequence_fasta_id: None,
COLUMNS.protocol: PROTOCOL.GENOME,
},
)
self.gev2 = BreakpointPair(
Breakpoint('1', 1),
Breakpoint('1', 10),
opposing_strands=True,
data={
COLUMNS.event_type: SVTYPE.DEL,
COLUMNS.call_method: CALL_METHOD.CONTIG,
COLUMNS.fusion_sequence_fasta_id: None,
COLUMNS.protocol: PROTOCOL.GENOME,
},
)
self.ust1 = PreTranscript(exons=[(1, 100), (301, 400), (501, 600)],
strand=STRAND.POS,
name='t1')
self.ust2 = PreTranscript(exons=[(1001, 1100), (1301, 1400),
(1501, 1600)],
strand=STRAND.POS,
name='t2')
self.distances = {
CALL_METHOD.CONTIG: 0,
CALL_METHOD.FLANK: 0,
CALL_METHOD.SPLIT: 10
}
self.TRANSCRIPTS = {
self.ust1.name: self.ust1,
self.ust2.name: self.ust2
}
def pos_splicing_pattern():
n = argparse.Namespace()
n.ex1 = Exon(100, 199, strand=STRAND.POS) # C
n.ex2 = Exon(500, 599, strand=STRAND.POS) # G
n.ex3 = Exon(1200, 1299, strand=STRAND.POS) # T
n.ex4 = Exon(1500, 1599, strand=STRAND.POS) # C
n.ex5 = Exon(1700, 1799, strand=STRAND.POS) # G
n.ex6 = Exon(2000, 2099, strand=STRAND.POS) # C
# introns: 99, 300, 600, 200, 100, ...
reference_sequence = 'a' * 99 + 'C' * 100 + 'a' * 300 + 'G' * 100
reference_sequence += 'a' * 600 + 'T' * 100 + 'a' * 200 + 'C' * 100
reference_sequence += 'a' * 100 + 'G' * 100 + 'a' * 200 + 'C' * 100
n.reference_sequence = reference_sequence
n.pre_transcript = PreTranscript(
exons=[n.ex1, n.ex2, n.ex3, n.ex4, n.ex5, n.ex6], strand=STRAND.POS)
return n
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
class TestTraverseTransRev(unittest.TestCase):
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_left_before_transcript(self):
gpos = self.trans_evidence.traverse(900, 500 - 1, ORIENT.LEFT)
self.assertEqual(Interval(401), gpos)
self.assertEqual(gpos,
GenomeEvidence.traverse(900, 500 - 1, ORIENT.LEFT))
def test_left_after_transcript(self):
gpos = self.trans_evidence.traverse(2200, 100, ORIENT.LEFT)
self.assertEqual(gpos, GenomeEvidence.traverse(2200, 100, ORIENT.LEFT))
self.assertEqual(Interval(2100), gpos)
def test_left_after_transcript2(self):
gpos = self.trans_evidence.traverse(1900, 500 - 1, ORIENT.LEFT)
self.assertEqual(Interval(901), gpos)
def test_left_within_transcript_exonic(self):
gpos = self.trans_evidence.traverse(1750, 200 - 1, ORIENT.LEFT)
self.assertEqual(Interval(1051), gpos)
def test_left_within_exon(self):
gpos = self.trans_evidence.traverse(1750, 20 - 1, ORIENT.LEFT)
self.assertEqual(1731, gpos.start)
self.assertEqual(1731, gpos.end)
def test_left_within_transcript_intronic(self):
gpos = self.trans_evidence.traverse(1600, 150 - 1, ORIENT.LEFT)
self.assertEqual(Interval(1451), gpos)
def test_right_before_transcript(self):
gpos = self.trans_evidence.traverse(500, 100 - 1, ORIENT.RIGHT)
self.assertEqual(Interval(599), gpos)
def test_right_before_transcript2(self):
gpos = self.trans_evidence.traverse(901, 500 - 1, ORIENT.RIGHT)
self.assertEqual(Interval(1900), gpos)
def test_right_after_transcript(self):
gpos = self.trans_evidence.traverse(2201, 100 - 1, ORIENT.RIGHT)
self.assertEqual(Interval(2300), gpos)
def test_right_within_transcript(self):
gpos = self.trans_evidence.traverse(1351, 100 - 1, ORIENT.RIGHT)
self.assertEqual(Interval(1750), gpos)
def test_right_within_exon(self):
gpos = self.trans_evidence.traverse(1351, 10 - 1, ORIENT.RIGHT)
self.assertEqual(Interval(1360), gpos)
def setUp(self):
self.pre_transcript = PreTranscript([(101, 200), (301, 400),
(501, 600)],
strand=STRAND.POS)
self.n_ust = PreTranscript([(101, 200), (301, 400), (501, 600)],
strand=STRAND.NEG)
def test_single_exon(self):
t = PreTranscript([(3, 4)], strand=STRAND.POS)
patt = t.generate_splicing_patterns()
assert len(patt) == 1
assert len(patt[0]) == 0
assert patt[0].splice_type == SPLICE_TYPE.NORMAL
def unspliced_transcript2():
return PreTranscript(
exons=[(1001, 1100), (1301, 1400), (1501, 1600)], strand=STRAND.POS, name='t2'
)
def test_single_exon(self):
t = PreTranscript([(3, 4)], strand=STRAND.POS)
patt = t.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(0, len(patt[0]))
self.assertEqual(SPLICE_TYPE.NORMAL, patt[0].splice_type)
class TestSplicingPatterns(unittest.TestCase):
def setUp(self):
self.setup_by_strand(STRAND.POS)
def setup_by_strand(self, strand):
self.ex1 = Exon(100, 199, strand=strand) # C
self.ex2 = Exon(500, 599, strand=strand) # G
self.ex3 = Exon(1200, 1299, strand=strand) # T
self.ex4 = Exon(1500, 1599, strand=strand) # C
self.ex5 = Exon(1700, 1799, strand=strand) # G
self.ex6 = Exon(2000, 2099, strand=strand) # C
# introns: 99, 300, 600, 200, 100, ...
reference_sequence = 'a' * 99 + 'C' * 100 + 'a' * 300 + 'G' * 100
reference_sequence += 'a' * 600 + 'T' * 100 + 'a' * 200 + 'C' * 100
reference_sequence += 'a' * 100 + 'G' * 100 + 'a' * 200 + 'C' * 100
self.reference_sequence = reference_sequence
self.pre_transcript = PreTranscript(exons=[self.ex1, self.ex2, self.ex3, self.ex4, self.ex5, self.ex6], strand=strand)
def test_single_exon(self):
t = PreTranscript([(3, 4)], strand=STRAND.POS)
patt = t.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(0, len(patt[0]))
self.assertEqual(SPLICE_TYPE.NORMAL, patt[0].splice_type)
def test_normal_pattern_pos(self):
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex2.start,
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.NORMAL, patt[0].splice_type)
def test_normal_pattern_neg(self):
self.setup_by_strand(STRAND.NEG)
self.assertTrue(self.pre_transcript.is_reverse)
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex2.start,
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
sorted([s.pos for s in patt[0]])
)
self.assertEqual(SPLICE_TYPE.NORMAL, patt[0].splice_type)
def test_abrogate_a_pos(self):
self.ex2.start_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(2, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.SKIP, patt[0].splice_type)
self.assertEqual(
[
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[1]]
)
self.assertEqual(SPLICE_TYPE.RETAIN, patt[1].splice_type)
def test_abrogate_a_neg(self):
self.setup_by_strand(STRAND.NEG)
self.ex2.start_splice_site.intact = False
patt = sorted(self.pre_transcript.generate_splicing_patterns())
self.assertEqual(2, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
sorted([s.pos for s in patt[0]])
)
self.assertEqual(SPLICE_TYPE.SKIP, patt[0].splice_type)
self.assertEqual(
[
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
sorted([s.pos for s in patt[1]])
)
self.assertEqual(SPLICE_TYPE.RETAIN, patt[1].splice_type)
def test_abrogate_a_last_exon(self):
self.ex6.start_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex2.start,
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.RETAIN, patt[0].splice_type)
def test_abrogate_d_first_exon(self):
self.ex1.end_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(
[
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.RETAIN, patt[0].splice_type)
def test_abrogate_ad(self):
self.ex2.start_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(2, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.SKIP, patt[0].splice_type)
self.assertEqual(
[
self.ex2.end, self.ex3.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[1]]
)
self.assertEqual(SPLICE_TYPE.RETAIN, patt[1].splice_type)
def test_abrogate_da(self):
self.ex2.end_splice_site.intact = False
self.ex3.start_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(1, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex2.start,
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.RETAIN, patt[0].splice_type)
def test_multiple_exons_or_multiple_introns_abrogate_ada(self):
self.ex2.start_splice_site.intact = False
self.ex2.end_splice_site.intact = False
self.ex3.start_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(2, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.MULTI_SKIP, patt[0].splice_type)
self.assertEqual(
[
self.ex3.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[1]]
)
self.assertEqual(SPLICE_TYPE.MULTI_RETAIN, patt[1].splice_type)
def test_multiple_exons_or_multiple_introns_abrogate_dad(self):
self.ex2.end_splice_site.intact = False
self.ex3.start_splice_site.intact = False
self.ex3.end_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(2, len(patt))
self.assertEqual(
[
self.ex1.end, self.ex2.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[0]]
)
self.assertEqual(SPLICE_TYPE.MULTI_RETAIN, patt[0].splice_type)
self.assertEqual(
[
self.ex1.end, self.ex4.start,
self.ex4.end, self.ex5.start,
self.ex5.end, self.ex6.start
],
[s.pos for s in patt[1]]
)
self.assertEqual(SPLICE_TYPE.MULTI_SKIP, patt[1].splice_type)
def test_complex(self):
self.ex2.end_splice_site.intact = False
self.ex4.end_splice_site.intact = False
patt = self.pre_transcript.generate_splicing_patterns()
self.assertEqual(4, len(patt))
self.assertTrue(SPLICE_TYPE.COMPLEX in [p.splice_type for p in patt])
def negative_transcript():
return PreTranscript([(101, 200), (301, 400), (501, 600)], strand=STRAND.NEG)
class TestTranscriptomeEvidenceWindow(unittest.TestCase):
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 transcriptome_window(self, breakpoint, transcripts=None):
if transcripts:
self.trans_evidence.overlapping_transcripts.update(transcripts)
return TranscriptomeEvidence.generate_window(self.trans_evidence,
breakpoint)
def genome_window(self, breakpoint):
return GenomeEvidence.generate_window(self.genome_evidence, breakpoint)
def test_before_start(self):
b = Breakpoint(chr='1', start=100, orient=ORIENT.RIGHT)
self.assertEqual(self.genome_window(b), self.transcriptome_window(b))
b = Breakpoint(chr='1', start=500, orient=ORIENT.RIGHT)
self.assertEqual(self.genome_window(b), self.transcriptome_window(b))
def test_after_end(self):
b = Breakpoint(chr='1', start=6000, orient=ORIENT.RIGHT)
self.assertEqual(self.genome_window(b), self.transcriptome_window(b))
def test_exonic_long_exon(self):
b = Breakpoint(chr='1', start=3200, orient=ORIENT.RIGHT)
self.assertEqual(self.genome_window(b), self.transcriptome_window(b))
def test_intronic_long_exon(self):
b = Breakpoint(chr='1', start=2970, orient=ORIENT.RIGHT)
self.assertEqual(self.genome_window(b), self.transcriptome_window(b))
def test_intronic_long_intron(self):
b = Breakpoint(chr='1', start=1800, orient=ORIENT.RIGHT)
print(self.genome_window(b))
self.assertEqual(Interval(1490, 2360), self.transcriptome_window(b))
def test_intronic_short_exon_right(self):
b = Breakpoint(chr='1', start=1690, orient=ORIENT.RIGHT)
print(self.genome_window(b))
self.assertEqual(Interval(1580, 3500), self.transcriptome_window(b))
def test_intronic_short_exon_left(self):
b = Breakpoint(chr='1', start=2200, orient=ORIENT.LEFT)
self.assertEqual(Interval(1440, 2310), self.transcriptome_window(b))
def test_multiple_transcripts(self):
# [(1001, 1100), (1401, 1500), (1701, 1750), (3001, 4000)])
b = Breakpoint(chr='1', start=1150, orient=ORIENT.RIGHT)
gene = self.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
self.assertEqual(
Interval(1040, 3160),
self.transcriptome_window(b, [self.pre_transcript, t2]))
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 positive_transcript():
return PreTranscript([(101, 200), (301, 400), (501, 600)], strand=STRAND.POS)
def unspliced_transcript1():
return PreTranscript(exons=[(1, 100), (301, 400), (501, 600)], strand=STRAND.POS, name='t1')