def test_only_CDS_overlap(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 1250
t2.end = 2000
t2.add_exons([(1250, 1560), (1801, 2000)])
t2.add_exons([(1401, 1560), (1801, 1850)], "CDS")
t2.finalize()
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
t2.strip_cds()
t2.finalized = False
t2.add_exons([(1461, 1560), (1801, 1850)], "CDS")
# No CDS overlap this time
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
class TranscriptTester(unittest.TestCase):
tr_gff = """Chr1 TAIR10 mRNA 5928 8737 . . . ID=AT1G01020.1;Parent=AT1G01020;Name=AT1G01020.1;Index=1
Chr1 TAIR10 exon 5928 8737 . . . Parent=AT1G01020.1"""
tr_lines = tr_gff.split("\n")
for pos, line in enumerate(tr_lines):
tr_lines[pos] = re.sub("\s+", "\t", line)
assert len(tr_lines[pos].split("\t")) == 9, line.split("\t")
tr_gff_lines = [Mikado.parsers.GFF.GffLine(line) for line in tr_lines]
for l in tr_gff_lines:
assert l.header is False
# print(l)
logger = create_null_logger("null")
def setUp(self):
"""Basic creation test."""
self.tr = Transcript()
self.tr.logger = self.logger
self.tr.chrom = "Chr1"
self.tr.source = "TAIR10"
self.tr.feature = "mRNA"
self.tr.start = 5928
self.tr.end = 8737
self.tr.strand = "+"
self.tr.add_exon((5928, 8737))
self.tr.score = None
self.tr.id, self.tr.parent, self.tr.name = "AT1G01020.1", "AT1G01020", "AT1G01020.1"
self.tr.add_exon((8571, 8666), "CDS")
self.tr.finalize()
self.orf = Mikado.parsers.bed12.BED12()
self.orf.chrom = self.tr.id
self.orf.start = 1
self.orf.end = self.tr.cdna_length
self.orf.name = self.tr.id
self.orf.strand = "+"
self.orf.score = 0
self.orf.thick_start = 8571 - 5928 + 1
self.orf.thick_end = 8666 - 5928 + 1
self.orf.block_count = 1
self.orf.blockSize = self.tr.cdna_length
self.orf.block_starts = 0
self.orf.has_start_codon = True
self.orf.has_stop_codon = True
self.orf.transcriptomic = True
self.assertFalse(self.orf.invalid, self.orf.invalid_reason)
self.assertEqual((self.orf.thick_end - self.orf.thick_start + 1) % 3, 0)
def test_invalid_inizialization(self):
with self.assertRaises(TypeError):
_ = Mikado.loci.Transcript(self.tr_gff_lines[1])
def test_basics(self):
self.assertEqual(self.tr.chrom, "Chr1")
self.assertEqual(self.tr.exon_num, 1)
self.assertEqual(self.tr.monoexonic, True)
self.assertEqual(self.tr.exon_num, len(self.tr.exons))
self.assertEqual(self.tr.start, 5928)
self.assertEqual(self.tr.end, 8737)
self.assertEqual(self.tr.exons,
[tuple([5928, 8737])],
self.tr.exons)
def test_cds(self):
"""Test the CDS features.
Note that in a single-exon transcript with no strand, start_codon and stop_codon are defined as False.
"""
self.tr.load_orfs([self.orf])
self.assertEqual(self.tr.combined_cds, self.tr.selected_cds)
self.assertEqual(self.tr.combined_cds,
[tuple([8571, 8666])],
self.tr.combined_cds)
self.assertEqual(self.tr.selected_cds_start, 8571)
self.assertEqual(self.tr.selected_cds_end, 8666)
self.assertEqual(self.tr.has_start_codon, True)
self.assertEqual(self.tr.has_stop_codon, True)
def test_equality(self):
new_transcript = self.tr.deepcopy()
self.assertTrue(new_transcript == self.tr)
new_transcript.strand = None
self.assertFalse(new_transcript == self.tr) # They have now a different strand
new_transcript.unfinalize()
new_transcript.strand = "+" # It becomes a multiexonic transcript, so it must have a strand
new_transcript.end = 9737
new_exon = Mikado.parsers.GFF.GffLine(self.tr_lines[-1])
new_exon.strand = "+"
new_exon.start = 9000
new_exon.end = 9737
new_transcript.add_exon(new_exon)
new_transcript.finalize()
self.assertTrue(new_transcript != self.tr)
def test_mono_finalising(self):
transcript_line = [line for line in self.tr_gff_lines if line.feature == "mRNA" ]
self.assertEqual(len(transcript_line), 1,
"\n".join([str(line) for line in self.tr_gff_lines]))
tr = Mikado.loci.Transcript(transcript_line[0])
exon_lines = [line for line in self.tr_gff_lines if
line.is_exon is True and "UTR" not in line.feature.upper()]
tr.add_exons(exon_lines)
tr.add_exon((8571, 8666), "CDS")
tr.finalize()
self.assertGreater(tr.three_utr_length, 0)
self.assertGreater(tr.five_utr_length, 0)
def test_invalid_transcript(self):
lines = """Chr1\tTAIR10\tmRNA\t5928\t8737\t.\t.\t.\tID=AT1G01020.1;Parent=AT1G01020;Name=AT1G01020.1;Index=1
Chr1\tTAIR10\tCDS\t8571\t7500\t.\t.\t0\tParent=AT1G01020.1;
Chr1\tTAIR10\tCDS\t7503\t8666\t.\t.\t0\tParent=AT1G01020.1;
Chr1\tTAIR10\texon\t5928\t8737\t.\t.\t.\tParent=AT1G01020.1"""
gff_lines = [Mikado.parsers.GFF.GffLine(line) for line in lines.split("\n")]
self.assertIsInstance(gff_lines[0], Mikado.parsers.GFF.GffLine)
checker = False
if gff_lines[0].feature.endswith("transcript") or "RNA" in gff_lines[0].feature.upper():
checker = True
self.assertTrue(checker)
self.assertTrue(gff_lines[0].is_transcript)
transcript = Mikado.loci.Transcript(gff_lines[0])
transcript.logger = self.logger
transcript.add_exons(gff_lines[1:])
with self.assertRaises(Mikado.exceptions.InvalidCDS):
Mikado.transcripts.transcript_methods.finalizing._check_cdna_vs_utr(transcript)
def test_utr(self):
self.assertEqual(self.tr.selected_internal_orf,
[("UTR", tuple([5928, 8570])),
("exon", tuple([5928, 8737])),
("CDS", tuple([8571, 8666]), 0),
("UTR", tuple([8667, 8737]))],
"Right: {0}\nFound{1}".format([("UTR", 5928, 8570), ("CDS", 8571, 8666), ("UTR", 8667, 8737)],
self.tr.selected_internal_orf))
self.assertEqual(self.tr.combined_utr, [tuple([5928, 8570]),
tuple([8667, 8737])])
self.assertEqual(self.tr.five_utr, [tuple([5928, 8570])],
self.tr.five_utr)
self.assertEqual(self.tr.three_utr, [tuple([8667, 8737])])
def test_utr_metrics(self):
"""Test for UTR exon num, start distance, etc."""
self.assertEqual(self.tr.five_utr_num, 1)
self.assertEqual(self.tr.three_utr_num, 1)
self.assertEqual(self.tr.five_utr_length, 8570 + 1 - 5928)
self.assertEqual(self.tr.three_utr_length, 8737 + 1 - 8667)
self.assertEqual(self.tr.selected_start_distance_from_tss,
8571 - 5928,
self.tr.selected_end_distance_from_tes)
self.assertEqual(self.tr.selected_end_distance_from_tes,
8737 - 8666,
(self.tr.selected_end_distance_from_tes, self.tr.strand))
def test_strip_cds(self):
self.tr.strip_cds()
self.assertEqual(self.tr.selected_cds_length, 0)
self.assertEqual(self.tr.three_utr, [])
self.assertEqual(self.tr.five_utr, [])
self.assertEqual(self.tr.selected_cds, [])
self.assertEqual(self.tr.selected_cds_start, None)
self.assertEqual(self.tr.selected_cds_end, None)
def test_remove_utr(self):
"""Test for CDS stripping. We remove the UTRs and verify that start/end have moved, no UTR is present, etc.
"""
self.tr.remove_utrs()
self.assertEqual(self.tr.selected_cds_start, self.tr.start)
self.assertEqual(self.tr.selected_cds_end, self.tr.end)
self.assertEqual(self.tr.three_utr, [])
self.assertEqual(self.tr.five_utr, [])
self.assertEqual(self.tr.combined_cds,
[tuple([8571, 8666])],
self.tr.combined_cds)
self.assertEqual(self.tr.combined_utr, [], self.tr.combined_utr)
def test_negative_orf(self):
"""Test loading a negative strand ORF onto a monoexonic transcript.
This should reverse the ORF."""
self.orf.strand = "-"
self.tr.strip_cds(strand_specific=False)
self.orf.has_stop_codon = False
self.tr.load_orfs([self.orf])
self.assertEqual(self.tr.strand, "-")
self.assertEqual(self.tr.selected_cds_start, 8666)
self.assertEqual(self.tr.selected_cds_end, 8571)
def test_introns(self):
self.assertEqual(self.tr.introns,
set([
]),
self.tr.introns
)
self.assertEqual(self.tr.combined_cds_introns,
set([
]),
self.tr.combined_cds_introns
)
self.assertEqual(self.tr.selected_cds_introns,
set([
]),
self.tr.selected_cds_introns
)
def testDoubleOrf(self):
"""Test to verify the introduction of multiple ORFs."""
self.tr.strip_cds()
self.tr.finalized = False
first_orf = Mikado.parsers.bed12.BED12()
first_orf.chrom = self.tr.id
first_orf.start = 1
first_orf.end = self.tr.cdna_length
first_orf.name = self.tr.id
first_orf.strand = "+"
first_orf.score = 0
first_orf.thick_start = 51
first_orf.thick_end = 398
first_orf.block_count = 1
first_orf.blockSize = self.tr.cdna_length
first_orf.block_sizes = [self.tr.cdna_length]
first_orf.block_starts = [0]
first_orf.rgb = 0
first_orf.has_start_codon = True
first_orf.has_stop_codon = True
first_orf.transcriptomic = True
self.assertFalse(first_orf.invalid)
# This should not be incorporated
second_orf = Mikado.parsers.bed12.BED12()
second_orf.chrom = self.tr.id
second_orf.start = 1
second_orf.end = self.tr.cdna_length
second_orf.name = "second"
second_orf.strand = "+"
second_orf.score = 0
second_orf.thick_start = 201
second_orf.thick_end = 410
second_orf.block_count = 1
second_orf.blockSize = self.tr.cdna_length
second_orf.block_sizes = [self.tr.cdna_length]
second_orf.block_starts = [0]
second_orf.rgb = 0
second_orf.has_start_codon = True
second_orf.has_stop_codon = True
second_orf.transcriptomic = True
self.assertFalse(second_orf.invalid)
self.assertTrue(Mikado.loci.Transcript.is_overlapping_cds(
first_orf, second_orf))
# This should be added
third_orf = Mikado.parsers.bed12.BED12()
third_orf.chrom = self.tr.id
third_orf.start = 1
third_orf.end = self.tr.cdna_length
third_orf.name = "third"
third_orf.strand = "+"
third_orf.score = 0
third_orf.thick_start = 501
third_orf.thick_end = 800
third_orf.block_count = 1
third_orf.blockSize = self.tr.cdna_length
third_orf.block_sizes = [self.tr.cdna_length]
third_orf.block_starts = [0]
third_orf.rgb = 0
third_orf.has_start_codon = True
third_orf.has_stop_codon = True
third_orf.transcriptomic = True
self.assertFalse(third_orf.invalid)
self.assertFalse(Mikado.loci.Transcript.is_overlapping_cds(first_orf, third_orf))
self.assertFalse(Mikado.loci.Transcript.is_overlapping_cds(second_orf, third_orf))
self.assertFalse(third_orf == second_orf)
self.assertFalse(first_orf == second_orf)
self.assertFalse(first_orf == third_orf)
candidates = [first_orf, second_orf, third_orf]
self.tr.logger = self.logger
self.tr.load_orfs([first_orf])
self.tr.load_orfs([second_orf])
self.tr.load_orfs([third_orf])
self.tr.load_orfs([first_orf, second_orf, third_orf])
self.assertTrue(self.tr.is_complete)
self.tr.finalize()
self.assertEqual(self.tr.number_internal_orfs, 2,
(self.tr.cdna_length, self.tr.selected_start_distance_from_tss,
self.tr.selected_end_distance_from_tes))
self.assertEqual(self.tr.combined_cds_length, 648)
self.assertEqual(self.tr.selected_cds_length, 348)
self.assertEqual(self.tr.number_internal_orfs, 2,
"\n".join([str(x) for x in self.tr.internal_orfs]))
new_transcripts = sorted(self.tr.split_by_cds())
self.assertEqual(len(new_transcripts), 2)
self.assertEqual(new_transcripts[0].three_utr_length, 0)
self.assertEqual(new_transcripts[1].five_utr_length, 0)
def testDoubleOrf_negative(self):
"""Test to verify the introduction of multiple ORFs."""
self.tr.strip_cds(strand_specific=False)
self.tr.finalized = False
first_orf = Mikado.parsers.bed12.BED12()
first_orf.chrom = self.tr.id
first_orf.start = 1
first_orf.end = self.tr.cdna_length
first_orf.name = self.tr.id
first_orf.strand = "-"
first_orf.score = 0
first_orf.thick_start = 51
first_orf.thick_end = 398
first_orf.block_count = 1
first_orf.blockSize = self.tr.cdna_length
first_orf.block_sizes = [self.tr.cdna_length]
first_orf.block_starts = [0]
first_orf.rgb = 0
first_orf.has_start_codon = True
first_orf.has_stop_codon = True
first_orf.transcriptomic = True
self.assertFalse(first_orf.invalid)
# This should not be incorporated
second_orf = Mikado.parsers.bed12.BED12()
second_orf.chrom = self.tr.id
second_orf.start = 1
second_orf.end = self.tr.cdna_length
second_orf.name = "second"
second_orf.strand = "-"
second_orf.score = 0
second_orf.thick_start = 201
second_orf.thick_end = 410
second_orf.block_count = 1
second_orf.blockSize = self.tr.cdna_length
second_orf.block_sizes = [self.tr.cdna_length]
second_orf.block_starts = [0]
second_orf.rgb = 0
second_orf.has_start_codon = True
second_orf.has_stop_codon = True
second_orf.transcriptomic = True
self.assertFalse(second_orf.invalid)
# self.assertTrue(Mikado.loci.Transcript.is_overlapping_cds(first_orf,
# second_orf))
# This should be added
third_orf = Mikado.parsers.bed12.BED12()
third_orf.chrom = self.tr.id
third_orf.start = 1
third_orf.end = self.tr.cdna_length
third_orf.name = "third"
third_orf.strand = "-"
third_orf.score = 0
third_orf.thick_start = 501
third_orf.thick_end = 800
third_orf.block_count = 1
third_orf.blockSize = self.tr.cdna_length
third_orf.block_sizes = [self.tr.cdna_length]
third_orf.block_starts = [0]
third_orf.rgb = 0
third_orf.has_start_codon = True
third_orf.has_stop_codon = True
third_orf.transcriptomic = True
self.assertFalse(third_orf.invalid)
self.assertFalse(
Mikado.loci.Transcript.is_overlapping_cds(
first_orf, third_orf))
self.assertFalse(
Mikado.loci.Transcript.is_overlapping_cds(
second_orf, third_orf))
self.assertFalse(third_orf == second_orf)
self.assertFalse(first_orf == second_orf)
self.assertFalse(first_orf == third_orf)
candidates = [first_orf, second_orf, third_orf]
# self.assertEqual(len(self.tr.find_overlapping_cds(candidates)), 2)
self.tr.logger = self.logger
self.tr.load_orfs(candidates)
self.assertTrue(self.tr.is_complete)
self.tr.finalize()
self.assertEqual(self.tr.number_internal_orfs, 2, (
self.tr.cdna_length,
self.tr.selected_start_distance_from_tss,
self.tr.selected_end_distance_from_tes))
# self.assertEqual(self.tr.combined_cds_length, 648)
self.assertEqual(self.tr.selected_cds_length, 348)
self.assertEqual(self.tr.number_internal_orfs, 2, "\n".join([str(x) for x in self.tr.internal_orfs]))
new_transcripts = sorted(self.tr.split_by_cds())
self.assertEqual(len(new_transcripts), 2)
self.assertEqual(new_transcripts[0].five_utr_length, 0)
self.assertEqual(new_transcripts[1].three_utr_length, 0)
def test_wrong_orf(self):
# This should be added
orf = Mikado.parsers.bed12.BED12()
orf.chrom = self.tr.id
orf.start = 1
orf.end = self.tr.cdna_length + 1
orf.name = "third"
orf.strand = "-"
orf.score = 0
orf.thick_start = 501
orf.thick_end = 800
orf.block_count = 1
orf.blockSize = self.tr.cdna_length
orf.block_sizes = [self.tr.cdna_length]
orf.block_starts = [0]
orf.rgb = 0
orf.has_start_codon = True
orf.has_stop_codon = True
orf.transcriptomic = True
self.assertFalse(orf.invalid)
self.tr.logger = self.logger
self.tr.strip_cds()
self.tr.strand = "+"
self.logger.setLevel("WARNING")
# self.tr.load_orfs([orf])
with self.assertLogs("null", level="DEBUG") as cm_out:
self.tr.load_orfs([orf])
self.assertFalse(self.tr.is_coding)
class TranscriptTester(unittest.TestCase):
tr_gff = """Chr1 TAIR10 mRNA 5928 8737 . . . ID=AT1G01020.1;Parent=AT1G01020;Name=AT1G01020.1;Index=1
Chr1 TAIR10 exon 5928 8737 . . . Parent=AT1G01020.1"""
tr_lines = tr_gff.split("\n")
for pos, line in enumerate(tr_lines):
tr_lines[pos] = re.sub("\s+", "\t", line)
assert len(tr_lines[pos].split("\t")) == 9, line.split("\t")
tr_gff_lines = [Mikado.parsers.GFF.GffLine(line) for line in tr_lines]
for l in tr_gff_lines:
assert l.header is False
# print(l)
logger = create_null_logger("null")
def setUp(self):
"""Basic creation test."""
self.tr = Transcript()
self.tr.logger = self.logger
self.tr.chrom = "Chr1"
self.tr.source = "TAIR10"
self.tr.feature = "mRNA"
self.tr.start = 5928
self.tr.end = 8737
self.tr.strand = "+"
self.tr.add_exon((5928, 8737))
self.tr.score = None
self.tr.id, self.tr.parent, self.tr.name = "AT1G01020.1", "AT1G01020", "AT1G01020.1"
self.tr.add_exon((8571, 8666), "CDS")
self.tr.finalize()
self.orf = Mikado.parsers.bed12.BED12()
self.orf.chrom = self.tr.id
self.orf.start = 1
self.orf.end = self.tr.cdna_length
self.orf.name = self.tr.id
self.orf.strand = "+"
self.orf.score = 0
self.orf.thick_start = 8571 - 5928 + 1
self.orf.thick_end = 8666 - 5928 + 1
self.orf.block_count = 1
self.orf.blockSize = self.tr.cdna_length
self.orf.block_starts = 0
self.orf.has_start_codon = True
self.orf.has_stop_codon = True
self.orf.transcriptomic = True
self.assertFalse(self.orf.invalid, self.orf.invalid_reason)
self.assertEqual((self.orf.thick_end - self.orf.thick_start + 1) % 3,
0)
def test_invalid_inizialization(self):
with self.assertRaises(TypeError):
_ = Mikado.loci.Transcript(self.tr_gff_lines[1])
def test_basics(self):
self.assertEqual(self.tr.chrom, "Chr1")
self.assertEqual(self.tr.exon_num, 1)
self.assertEqual(self.tr.monoexonic, True)
self.assertEqual(self.tr.exon_num, len(self.tr.exons))
self.assertEqual(self.tr.start, 5928)
self.assertEqual(self.tr.end, 8737)
self.assertEqual(self.tr.exons, [tuple([5928, 8737])], self.tr.exons)
def test_cds(self):
"""Test the CDS features.
Note that in a single-exon transcript with no strand, start_codon and stop_codon are defined as False.
"""
self.tr.load_orfs([self.orf])
self.assertEqual(self.tr.combined_cds, self.tr.selected_cds)
self.assertEqual(self.tr.combined_cds, [tuple([8571, 8666])],
self.tr.combined_cds)
self.assertEqual(self.tr.selected_cds_start, 8571)
self.assertEqual(self.tr.selected_cds_end, 8666)
self.assertEqual(self.tr.has_start_codon, True)
self.assertEqual(self.tr.has_stop_codon, True)
def test_equality(self):
new_transcript = self.tr.deepcopy()
self.assertTrue(new_transcript == self.tr)
new_transcript.strand = None
self.assertFalse(
new_transcript == self.tr) # They have now a different strand
new_transcript.unfinalize()
new_transcript.strand = "+" # It becomes a multiexonic transcript, so it must have a strand
new_transcript.end = 9737
new_exon = Mikado.parsers.GFF.GffLine(self.tr_lines[-1])
new_exon.strand = "+"
new_exon.start = 9000
new_exon.end = 9737
new_transcript.add_exon(new_exon)
new_transcript.finalize()
self.assertTrue(new_transcript != self.tr)
def test_mono_finalising(self):
transcript_line = [
line for line in self.tr_gff_lines if line.feature == "mRNA"
]
self.assertEqual(len(transcript_line), 1,
"\n".join([str(line) for line in self.tr_gff_lines]))
tr = Mikado.loci.Transcript(transcript_line[0])
exon_lines = [
line for line in self.tr_gff_lines
if line.is_exon is True and "UTR" not in line.feature.upper()
]
tr.add_exons(exon_lines)
tr.add_exon((8571, 8666), "CDS")
tr.finalize()
self.assertGreater(tr.three_utr_length, 0)
self.assertGreater(tr.five_utr_length, 0)
def test_invalid_transcript(self):
lines = """Chr1\tTAIR10\tmRNA\t5928\t8737\t.\t.\t.\tID=AT1G01020.1;Parent=AT1G01020;Name=AT1G01020.1;Index=1
Chr1\tTAIR10\tCDS\t8571\t7500\t.\t.\t0\tParent=AT1G01020.1;
Chr1\tTAIR10\tCDS\t7503\t8666\t.\t.\t0\tParent=AT1G01020.1;
Chr1\tTAIR10\texon\t5928\t8737\t.\t.\t.\tParent=AT1G01020.1"""
gff_lines = [
Mikado.parsers.GFF.GffLine(line) for line in lines.split("\n")
]
self.assertIsInstance(gff_lines[0], Mikado.parsers.GFF.GffLine)
checker = False
if gff_lines[0].feature.endswith(
"transcript") or "RNA" in gff_lines[0].feature.upper():
checker = True
self.assertTrue(checker)
self.assertTrue(gff_lines[0].is_transcript)
transcript = Mikado.loci.Transcript(gff_lines[0])
transcript.logger = self.logger
transcript.add_exons(gff_lines[1:])
with self.assertRaises(Mikado.exceptions.InvalidCDS):
Mikado.loci.transcript_methods.finalizing._check_cdna_vs_utr(
transcript)
def test_utr(self):
self.assertEqual(
self.tr.selected_internal_orf, [("UTR", tuple([5928, 8570])),
("exon", tuple([5928, 8737])),
("CDS", tuple([8571, 8666]), 0),
("UTR", tuple([8667, 8737]))],
"Right: {0}\nFound{1}".format([("UTR", 5928, 8570),
("CDS", 8571, 8666),
("UTR", 8667, 8737)],
self.tr.selected_internal_orf))
self.assertEqual(
self.tr.combined_utr,
[tuple([5928, 8570]), tuple([8667, 8737])])
self.assertEqual(self.tr.five_utr, [tuple([5928, 8570])],
self.tr.five_utr)
self.assertEqual(self.tr.three_utr, [tuple([8667, 8737])])
def test_utr_metrics(self):
"""Test for UTR exon num, start distance, etc."""
self.assertEqual(self.tr.five_utr_num, 1)
self.assertEqual(self.tr.three_utr_num, 1)
self.assertEqual(self.tr.five_utr_length, 8570 + 1 - 5928)
self.assertEqual(self.tr.three_utr_length, 8737 + 1 - 8667)
self.assertEqual(self.tr.selected_start_distance_from_tss, 8571 - 5928,
self.tr.selected_end_distance_from_tes)
self.assertEqual(
self.tr.selected_end_distance_from_tes, 8737 - 8666,
(self.tr.selected_end_distance_from_tes, self.tr.strand))
def test_strip_cds(self):
self.tr.strip_cds()
self.assertEqual(self.tr.selected_cds_length, 0)
self.assertEqual(self.tr.three_utr, [])
self.assertEqual(self.tr.five_utr, [])
self.assertEqual(self.tr.selected_cds, [])
self.assertEqual(self.tr.selected_cds_start, None)
self.assertEqual(self.tr.selected_cds_end, None)
def test_remove_utr(self):
"""Test for CDS stripping. We remove the UTRs and verify that start/end have moved, no UTR is present, etc.
"""
self.tr.remove_utrs()
self.assertEqual(self.tr.selected_cds_start, self.tr.start)
self.assertEqual(self.tr.selected_cds_end, self.tr.end)
self.assertEqual(self.tr.three_utr, [])
self.assertEqual(self.tr.five_utr, [])
self.assertEqual(self.tr.combined_cds, [tuple([8571, 8666])],
self.tr.combined_cds)
self.assertEqual(self.tr.combined_utr, [], self.tr.combined_utr)
def test_negative_orf(self):
"""Test loading a negative strand ORF onto a monoexonic transcript.
This should reverse the ORF."""
self.orf.strand = "-"
self.tr.strip_cds(strand_specific=False)
self.orf.has_stop_codon = False
self.tr.load_orfs([self.orf])
self.assertEqual(self.tr.strand, "-")
self.assertEqual(self.tr.selected_cds_start, 8666)
self.assertEqual(self.tr.selected_cds_end, 8571)
def test_introns(self):
self.assertEqual(self.tr.introns, set([]), self.tr.introns)
self.assertEqual(self.tr.combined_cds_introns, set([]),
self.tr.combined_cds_introns)
self.assertEqual(self.tr.selected_cds_introns, set([]),
self.tr.selected_cds_introns)
def testDoubleOrf(self):
"""Test to verify the introduction of multiple ORFs."""
self.tr.strip_cds()
self.tr.finalized = False
first_orf = Mikado.parsers.bed12.BED12()
first_orf.chrom = self.tr.id
first_orf.start = 1
first_orf.end = self.tr.cdna_length
first_orf.name = self.tr.id
first_orf.strand = "+"
first_orf.score = 0
first_orf.thick_start = 51
first_orf.thick_end = 398
first_orf.block_count = 1
first_orf.blockSize = self.tr.cdna_length
first_orf.block_sizes = [self.tr.cdna_length]
first_orf.block_starts = [0]
first_orf.rgb = 0
first_orf.has_start_codon = True
first_orf.has_stop_codon = True
first_orf.transcriptomic = True
self.assertFalse(first_orf.invalid)
# This should not be incorporated
second_orf = Mikado.parsers.bed12.BED12()
second_orf.chrom = self.tr.id
second_orf.start = 1
second_orf.end = self.tr.cdna_length
second_orf.name = "second"
second_orf.strand = "+"
second_orf.score = 0
second_orf.thick_start = 201
second_orf.thick_end = 410
second_orf.block_count = 1
second_orf.blockSize = self.tr.cdna_length
second_orf.block_sizes = [self.tr.cdna_length]
second_orf.block_starts = [0]
second_orf.rgb = 0
second_orf.has_start_codon = True
second_orf.has_stop_codon = True
second_orf.transcriptomic = True
self.assertFalse(second_orf.invalid)
self.assertTrue(
Mikado.loci.Transcript.is_overlapping_cds(first_orf, second_orf))
# This should be added
third_orf = Mikado.parsers.bed12.BED12()
third_orf.chrom = self.tr.id
third_orf.start = 1
third_orf.end = self.tr.cdna_length
third_orf.name = "third"
third_orf.strand = "+"
third_orf.score = 0
third_orf.thick_start = 501
third_orf.thick_end = 800
third_orf.block_count = 1
third_orf.blockSize = self.tr.cdna_length
third_orf.block_sizes = [self.tr.cdna_length]
third_orf.block_starts = [0]
third_orf.rgb = 0
third_orf.has_start_codon = True
third_orf.has_stop_codon = True
third_orf.transcriptomic = True
self.assertFalse(third_orf.invalid)
self.assertFalse(
Mikado.loci.Transcript.is_overlapping_cds(first_orf, third_orf))
self.assertFalse(
Mikado.loci.Transcript.is_overlapping_cds(second_orf, third_orf))
self.assertFalse(third_orf == second_orf)
self.assertFalse(first_orf == second_orf)
self.assertFalse(first_orf == third_orf)
candidates = [first_orf, second_orf, third_orf]
self.tr.logger = self.logger
self.tr.load_orfs([first_orf])
self.tr.load_orfs([second_orf])
self.tr.load_orfs([third_orf])
self.tr.load_orfs([first_orf, second_orf, third_orf])
self.assertTrue(self.tr.is_complete)
self.tr.finalize()
self.assertEqual(
self.tr.number_internal_orfs, 2,
(self.tr.cdna_length, self.tr.selected_start_distance_from_tss,
self.tr.selected_end_distance_from_tes))
self.assertEqual(self.tr.combined_cds_length, 648)
self.assertEqual(self.tr.selected_cds_length, 348)
self.assertEqual(self.tr.number_internal_orfs, 2,
"\n".join([str(x) for x in self.tr.internal_orfs]))
new_transcripts = sorted(self.tr.split_by_cds())
self.assertEqual(len(new_transcripts), 2)
self.assertEqual(new_transcripts[0].three_utr_length, 0)
self.assertEqual(new_transcripts[1].five_utr_length, 0)
def testDoubleOrf_negative(self):
"""Test to verify the introduction of multiple ORFs."""
self.tr.strip_cds(strand_specific=False)
self.tr.finalized = False
first_orf = Mikado.parsers.bed12.BED12()
first_orf.chrom = self.tr.id
first_orf.start = 1
first_orf.end = self.tr.cdna_length
first_orf.name = self.tr.id
first_orf.strand = "-"
first_orf.score = 0
first_orf.thick_start = 51
first_orf.thick_end = 398
first_orf.block_count = 1
first_orf.blockSize = self.tr.cdna_length
first_orf.block_sizes = [self.tr.cdna_length]
first_orf.block_starts = [0]
first_orf.rgb = 0
first_orf.has_start_codon = True
first_orf.has_stop_codon = True
first_orf.transcriptomic = True
self.assertFalse(first_orf.invalid)
# This should not be incorporated
second_orf = Mikado.parsers.bed12.BED12()
second_orf.chrom = self.tr.id
second_orf.start = 1
second_orf.end = self.tr.cdna_length
second_orf.name = "second"
second_orf.strand = "-"
second_orf.score = 0
second_orf.thick_start = 201
second_orf.thick_end = 410
second_orf.block_count = 1
second_orf.blockSize = self.tr.cdna_length
second_orf.block_sizes = [self.tr.cdna_length]
second_orf.block_starts = [0]
second_orf.rgb = 0
second_orf.has_start_codon = True
second_orf.has_stop_codon = True
second_orf.transcriptomic = True
self.assertFalse(second_orf.invalid)
# self.assertTrue(Mikado.loci.Transcript.is_overlapping_cds(first_orf,
# second_orf))
# This should be added
third_orf = Mikado.parsers.bed12.BED12()
third_orf.chrom = self.tr.id
third_orf.start = 1
third_orf.end = self.tr.cdna_length
third_orf.name = "third"
third_orf.strand = "-"
third_orf.score = 0
third_orf.thick_start = 501
third_orf.thick_end = 800
third_orf.block_count = 1
third_orf.blockSize = self.tr.cdna_length
third_orf.block_sizes = [self.tr.cdna_length]
third_orf.block_starts = [0]
third_orf.rgb = 0
third_orf.has_start_codon = True
third_orf.has_stop_codon = True
third_orf.transcriptomic = True
self.assertFalse(third_orf.invalid)
self.assertFalse(
Mikado.loci.Transcript.is_overlapping_cds(first_orf, third_orf))
self.assertFalse(
Mikado.loci.Transcript.is_overlapping_cds(second_orf, third_orf))
self.assertFalse(third_orf == second_orf)
self.assertFalse(first_orf == second_orf)
self.assertFalse(first_orf == third_orf)
candidates = [first_orf, second_orf, third_orf]
# self.assertEqual(len(self.tr.find_overlapping_cds(candidates)), 2)
self.tr.logger = self.logger
self.tr.load_orfs(candidates)
self.assertTrue(self.tr.is_complete)
self.tr.finalize()
self.assertEqual(
self.tr.number_internal_orfs, 2,
(self.tr.cdna_length, self.tr.selected_start_distance_from_tss,
self.tr.selected_end_distance_from_tes))
# self.assertEqual(self.tr.combined_cds_length, 648)
self.assertEqual(self.tr.selected_cds_length, 348)
self.assertEqual(self.tr.number_internal_orfs, 2,
"\n".join([str(x) for x in self.tr.internal_orfs]))
new_transcripts = sorted(self.tr.split_by_cds())
self.assertEqual(len(new_transcripts), 2)
self.assertEqual(new_transcripts[0].five_utr_length, 0)
self.assertEqual(new_transcripts[1].three_utr_length, 0)
def test_wrong_orf(self):
# This should be added
orf = Mikado.parsers.bed12.BED12()
orf.chrom = self.tr.id
orf.start = 1
orf.end = self.tr.cdna_length + 1
orf.name = "third"
orf.strand = "-"
orf.score = 0
orf.thick_start = 501
orf.thick_end = 800
orf.block_count = 1
orf.blockSize = self.tr.cdna_length
orf.block_sizes = [self.tr.cdna_length]
orf.block_starts = [0]
orf.rgb = 0
orf.has_start_codon = True
orf.has_stop_codon = True
orf.transcriptomic = True
self.assertFalse(orf.invalid)
self.tr.logger = self.logger
self.tr.strip_cds()
self.tr.strand = "+"
self.logger.setLevel("WARNING")
# self.tr.load_orfs([orf])
with self.assertLogs("null", level="DEBUG") as cm_out:
self.tr.load_orfs([orf])
self.assertFalse(self.tr.is_coding)
class TestMetricsEndDistances(unittest.TestCase):
logger = create_default_logger("End")
logger.setLevel("ERROR")
def setUp(self):
self.tr = Transcript()
self.tr.logger = self.logger
self.tr.start = 101
self.tr.end = 10000
self.tr.add_exons([(101, 300),
(501, 800),
(1001, 1200),
(1301, 2000),
(3501, 5000),
(5501, 6000),
(6201, 7000),
(7301, 7700),
(8201, 9000),
(9101, 9300),
(9501, 9700),
(9801, 10000)])
self.tr.id = "test1"
self.tr.parent = "test1.gene"
def test_end_positive(self):
self.tr.strand = "+"
cds = [(1161, 1200), # 40 % 3 == 1
(1301, 2000), # 700 % 3 == 1
(3501, 5000), # 1500 % 3 == 0
(5501, 6000), # 500 % 3 == 2
(6201, 7000), # 800 % 3 == 2
(7301, 7700), # 400 % 3 == 1
(8201, 9000), # 800 % 3 == 2
(9101, 9130)]
self.tr.add_exons(cds, features="CDS")
self.tr.finalize()
self.assertEqual(self.tr.selected_cds_end,
self.tr.combined_cds_end)
self.assertEqual(self.tr.selected_cds_end,
9130)
self.assertEqual(self.tr.end_distance_from_junction,
(9300 - 9131 + 1) + (9700 - 9501 + 1)
)
self.assertEqual(self.tr.end_distance_from_tes,
(9300 - 9131 + 1) + (9700 - 9501 + 1) + (10000 - 9801 + 1)
)
self.tr.strip_cds()
self.assertEqual(len(self.tr.internal_orfs), 0, self.tr.internal_orfs)
self.tr.finalized = False
cds = [(1161, 1200), # 40 % 3 == 1
(1301, 2000), # 700 % 3 == 1
(3501, 5000), # 1500 % 3 == 0
(5501, 6000), # 500 % 3 == 2
(6201, 7000), # 800 % 3 == 2
(7301, 7700), # 400 % 3 == 1
(8201, 9000), # 800 % 3 == 2
(9101, 9300), # 200 % 3 == 2
(9501, 9690) # 190 % 3 == 1
]
self.tr.add_exons(cds, features="CDS")
self.tr.finalize()
self.assertEqual(self.tr.combined_cds_end,
9690)
self.assertEqual(self.tr.selected_cds_end,
self.tr.combined_cds_end)
self.assertEqual(self.tr.end_distance_from_junction,
(9700 - 9691 + 1)
)
self.assertEqual(self.tr.end_distance_from_tes,
(9700 - 9691 + 1) + (10000 - 9801 + 1)
)
self.tr.strip_cds()
self.assertEqual(self.tr.combined_cds_end,
self.tr.selected_cds_end,
self.tr.combined_cds)
self.assertEqual(self.tr.combined_cds_end,
None,
self.tr.combined_cds_end)
self.tr.finalized = False
cds = [(1161, 1200), # 40 % 3 == 1
(1301, 2000), # 700 % 3 == 1
(3501, 5000), # 1500 % 3 == 0
(5501, 6000), # 500 % 3 == 2
(6201, 7000), # 800 % 3 == 2
(7301, 7700), # 400 % 3 == 1
(8201, 9000), # 800 % 3 == 2
(9101, 9300), # 200 % 3 == 2
(9501, 9700), # 200 % 3 == 2
(9801, 9820), # 20 % 2 == 2
]
self.tr.add_exons(cds, features="CDS")
self.tr.finalize()
self.assertEqual(self.tr.combined_cds_end,
9820)
self.assertEqual(self.tr.selected_cds_end,
self.tr.combined_cds_end)
self.assertEqual(self.tr.end_distance_from_tes,
180)
self.assertEqual(self.tr.end_distance_from_junction,
0)
def test_end_negative(self):
self.tr.strand = "-"
# self.tr.add_exons([(101, 300),
# (501, 800),
# (1001, 1200),
# (1301, 2000),
# (3501, 5000),
# (5501, 6000),
# (6201, 7000),
# (7301, 7700),
# (8201, 9000),
# (9101, 9300),
# (9501, 9700),
# (9801, 10000)])
cds = [(1161, 1200), # 40 % 3 == 1
(1301, 2000), # 700 % 3 == 1
(3501, 5000), # 1500 % 3 == 0
(5501, 6000), # 500 % 3 == 2
(6201, 7000), # 800 % 3 == 2
(7301, 7700), # 400 % 3 == 1
(8201, 9000), # 800 % 3 == 2
(9101, 9130)]
self.assertEqual(sum(x[1] - x[0] + 1 for x in cds) % 3, 0)
self.tr.add_exons(cds, features="CDS")
self.tr.finalize()
self.assertTrue(self.tr.is_coding)
self.assertEqual(self.tr.selected_cds_end,
self.tr.combined_cds_end)
self.assertEqual(self.tr.selected_cds_end,
1161)
self.assertEqual(self.tr.end_distance_from_junction,
(1161-1001) + (800-501+1),
(self.tr.end_distance_from_junction,
(1161-1001) + (800-501+1))
)
self.assertEqual(self.tr.end_distance_from_tes,
self.tr.end_distance_from_junction + (300 - 101 + 1),
(self.tr.end_distance_from_tes,
self.tr.end_distance_from_junction + (300 - 101 + 1))
)
self.tr.strip_cds()
self.assertEqual(len(self.tr.internal_orfs), 0, self.tr.internal_orfs)
self.tr.finalized = False
cds = [(721, 800),
(1001, 1200), # 200 % 3 == 2
(1301, 2000), # 700 % 3 == 1
(3501, 5000), # 1500 % 3 == 0
(5501, 6000), # 500 % 3 == 2
(6201, 7000), # 800 % 3 == 2
(7301, 7700), # 400 % 3 == 1
(8201, 9000), # 800 % 3 == 2
(9101, 9130), # 200 % 3 == 2
]
self.tr.add_exons(cds, features="CDS")
self.tr.finalize()
self.assertEqual(self.tr.combined_cds_end,
721)
self.assertEqual(self.tr.selected_cds_end,
self.tr.combined_cds_end)
self.assertEqual(self.tr.end_distance_from_junction,
(721-501),
(self.tr.end_distance_from_junction, (721-501))
)
self.assertEqual(self.tr.end_distance_from_tes,
self.tr.end_distance_from_junction + (300 - 101 + 1),
(self.tr.end_distance_from_tes,
self.tr.end_distance_from_junction + (300 - 101 + 1))
)
self.tr.strip_cds()
self.assertEqual(self.tr.combined_cds_end,
self.tr.selected_cds_end,
self.tr.combined_cds)
self.assertEqual(self.tr.combined_cds_end,
None,
self.tr.combined_cds_end)
self.tr.finalized = False
cds = [(161, 300), # 140 % 3 == 2
(501, 800), # 300 % 3 == 0
(1001, 1200), # 200 % 3 == 2
(1301, 2000), # 700 % 3 == 1
(3501, 5000), # 1500 % 3 == 0
(5501, 6000), # 500 % 3 == 2
(6201, 7000), # 800 % 3 == 2
(7301, 7700), # 400 % 3 == 1
(8201, 9000), # 800 % 3 == 2
(9101, 9130), # 30 % 3 == 0
]
self.assertEqual(sum((_[1] - _[0] +1) % 3 for _ in cds ) % 3, 0)
self.tr.logger = self.logger
self.tr.add_exons(cds, features="CDS")
self.tr.finalize()
self.assertEqual(self.tr.combined_cds_end,
161)
self.assertEqual(self.tr.selected_cds_end,
self.tr.combined_cds_end)
self.assertEqual(self.tr.end_distance_from_tes,
60)
self.assertEqual(self.tr.end_distance_from_junction,
0)
class MonoHolderTester(unittest.TestCase):
logger = create_default_logger("MonoHolderTester")
def setUp(self):
self.conf = dict()
self.t1 = Transcript()
self.t1.chrom = "Chr1"
self.t1.strand = "+"
self.t1.score = 20
self.t1.id = "G1.1"
self.t1.parent = "G1"
self.t1.start = 101
self.t1.end = 1500
self.t1.add_exons([(101, 500), (601, 700), (1001, 1300), (1401, 1500)],
"exon")
self.t1.add_exons([(401, 500), (601, 700), (1001, 1300), (1401, 1440)],
"CDS")
self.t1.finalize()
def testCdsOverlap(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 101
t2.end = 1600
t2.add_exons([(101, 500), (601, 700), (1001, 1300), (1401, 1460),
(1501, 1600)], "exon")
t2.add_exons([(401, 500), (601, 700), (1001, 1300), (1401, 1440)],
"CDS")
t2.finalize()
self.assertTrue(MonosublocusHolder.is_intersecting(self.t1, t2))
def test_intronMatch(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 101
t2.end = 1600
t2.add_exons([(101, 500), (601, 700), (1001, 1320), (1451, 1460),
(1501, 1600)], "exon")
t2.add_exons([(401, 500), (601, 700), (1001, 1320), (1451, 1460),
(1501, 1510)], "CDS")
t2.finalize()
self.assertTrue(self.t1.is_coding)
self.assertTrue(t2.is_coding)
self.assertTrue(
MonosublocusHolder.is_intersecting(self.t1, t2,
logger=self.logger))
self.assertTrue(
MonosublocusHolder.is_intersecting(self.t1,
t2,
cds_only=True,
logger=self.logger))
def test_intronOverlap(self):
self.t1.strip_cds()
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 101
t2.end = 1470
t2.add_exons([(101, 510), (601, 700), (960, 1350), (1420, 1470)])
t2.finalize()
self.assertTrue(MonosublocusHolder.is_intersecting(self.t1, t2))
def test_noIntronOverlap(self):
self.t1.strip_cds()
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 1250
t2.end = 2000
t2.add_exons([(1250, 1560), (1800, 2000)])
t2.finalize()
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
def test_noCDSOverlap(self):
self.t1.strip_cds()
self.assertEqual(self.t1.combined_cds_introns, set())
self.t1.finalized = False
self.t1.add_exons([(401, 500), (601, 700), (1001, 1100)], "CDS")
self.t1.finalize()
t2 = Transcript()
t2.logger = self.logger
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 101
t2.end = 1470
t2.add_exons([(101, 510), (601, 700), (960, 1350), (1421, 1470)])
t2.add_exons([(1201, 1350), (1421, 1450)], "CDS")
t2.finalize()
self.assertTrue(self.t1.is_coding)
self.assertTrue(t2.is_coding)
self.assertGreaterEqual(
0,
overlap((self.t1.combined_cds_start, self.t1.combined_cds_end),
(t2.combined_cds_start, t2.combined_cds_end)),
[(self.t1.combined_cds_start, self.t1.combined_cds_end),
(t2.combined_cds_start, t2.combined_cds_end)])
self.assertTrue(
MonosublocusHolder.is_intersecting(self.t1, t2,
logger=self.logger))
self.assertFalse(
MonosublocusHolder.is_intersecting(self.t1,
t2,
cds_only=True,
logger=self.logger))
def test_only_CDS_overlap(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 1250
t2.end = 2000
t2.add_exons([(1250, 1560), (1801, 2000)])
t2.add_exons([(1401, 1560), (1801, 1850)], "CDS")
t2.finalize()
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
t2.strip_cds()
t2.finalized = False
t2.add_exons([(1461, 1560), (1801, 1850)], "CDS")
# No CDS overlap this time
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
def test_no_overlap(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G2.1"
t2.parent = "G2"
t2.start = 1600
t2.end = 2000
t2.add_exons([(1600, 1700), (1801, 2000)])
t2.add_exons([(1661, 1700), (1801, 1850)], "CDS")
t2.finalize()
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
def test_same_id(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 1
t2.id = "G1.1"
t2.parent = "G1"
t2.start = 1250
t2.end = 2000
t2.add_exons([(1250, 1560), (1801, 2000)])
t2.add_exons([(1401, 1560), (1801, 1850)], "CDS")
t2.finalize()
# This fails because they have the same ID
self.assertFalse(MonosublocusHolder.is_intersecting(self.t1, t2))
