def test_wrong_cds(self):
transcript = Transcript()
transcript.chrom = "15"
transcript.source = "protein_coding"
transcript.start = 47631264
transcript.end = 48051999
exons = [(47631264, 47631416), (47704590, 47704669),
(47762671, 47762742), (47893062, 47893093),
(47895572, 47895655), (48051942, 48051999)]
transcript.strand = "+"
transcript.add_exons(exons)
transcript.id = "ENST00000560636"
transcript.parent = "ENSG00000137872"
cds_line = "\t".join([
"15", "protein_coding", "CDS", "48051996", "48051996", ".", "+",
"0", "ID=ENST00000560636.cds1;Parent=ENST00000560636"
])
cds_line = GffLine(cds_line)
transcript.add_exon(cds_line)
logger = Mikado.utilities.log_utils.create_null_logger()
transcript.logger = logger
with self.assertLogs("null", level="WARNING"):
transcript.finalize()
trimmed = trim_coding(transcript, logger, max_length=50)
self.assertEqual(trimmed.start, 47631366)
self.assertEqual(trimmed.end, 48051992)
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_coding_positive(self):
tr = Transcript()
tr.chrom = "Chr1"
tr.start = 101
tr.end = 2000
tr.strand = "+"
tr.add_exons([(101, 300),
(1701, 2000)])
tr.add_exons([(101, 300),
(1701, 2000)], features="CDS")
tr.id = "test1"
tr.parent = "gene1"
gff = tr.format("gff", with_introns=True)
self.maxDiff = None
res = """Chr1\tMikado\tmRNA\t101\t2000\t.\t+\t.\tID=test1;Parent=gene1;Name=test1
Chr1\tMikado\tCDS\t101\t300\t.\t+\t0\tID=test1.CDS1;Parent=test1
Chr1\tMikado\texon\t101\t300\t.\t+\t.\tID=test1.exon1;Parent=test1
Chr1\tMikado\tintron\t301\t1700\t.\t+\t.\tID=test1.intron1;Parent=test1
Chr1\tMikado\tCDS\t1701\t2000\t.\t+\t1\tID=test1.CDS2;Parent=test1
Chr1\tMikado\texon\t1701\t2000\t.\t+\t.\tID=test1.exon2;Parent=test1"""
self.assertEqual(gff,
res,
"++++\n\n" + "\n+++\n".join([gff, res]))
gtf = tr.format("gtf", with_introns=True)
res = """Chr1\tMikado\tmRNA\t101\t2000\t.\t+\t.\tgene_id "gene1"; transcript_id "test1"; Name "test1";
Chr1\tMikado\tCDS\t101\t300\t.\t+\t0\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t101\t300\t.\t+\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\tintron\t301\t1700\t.\t+\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\tCDS\t1701\t2000\t.\t+\t2\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t1701\t2000\t.\t+\t.\tgene_id "gene1"; transcript_id "test1";"""
self.assertEqual(gtf, res,
"++++\n\n" + "\n+++\n".join([gtf, res]))
def test_non_coding_negative(self):
tr = Transcript()
tr.chrom = "Chr1"
tr.start = 101
tr.end = 2000
tr.strand = "-"
tr.add_exons([(101, 300), (1701, 2000)])
tr.id = "test1"
tr.parent = "gene1"
tr.finalize()
gff = tr.format("gff", with_introns=True)
self.maxDiff = None
res = """Chr1\tMikado\ttranscript\t101\t2000\t.\t-\t.\tID=test1;Parent=gene1
Chr1\tMikado\texon\t101\t300\t.\t-\t.\tID=test1.exon1;Parent=test1
Chr1\tMikado\tintron\t301\t1700\t.\t-\t.\tID=test1.intron1;Parent=test1
Chr1\tMikado\texon\t1701\t2000\t.\t-\t.\tID=test1.exon2;Parent=test1"""
self.assertEqual(gff, res, "++++\n\n" + "\n+++\n".join([gff, res]))
gtf = tr.format("gtf", with_introns=True)
res = """Chr1\tMikado\ttranscript\t101\t2000\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t101\t300\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\tintron\t301\t1700\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t1701\t2000\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";"""
self.assertEqual(gtf, res, "++++\n\n" + "\n+++\n".join([gtf, res]))
def test_noncoding(self):
transcript = Transcript()
transcript.chrom = "Chr1"
transcript.source = "test"
transcript.start = 10000
transcript.end = 20000
exons = [(10000, 11500), (12000, 13000), (15000, 18000),
(19000, 20000)]
transcript.add_exons(exons)
transcript.strand = "+"
transcript.finalize()
logger = Mikado.utilities.log_utils.create_null_logger("correct_cds2")
copied = transcript.deepcopy()
trimmed = trim_noncoding(copied, max_length=50)
self.assertEqual(trimmed.start, 11450)
self.assertEqual(trimmed.end, 19050)
copied = transcript.deepcopy()
trimmed = trim_noncoding(copied, max_length=200)
self.assertEqual(trimmed.start, 11300)
self.assertEqual(trimmed.end, 19200)
def test_coding_negative(self):
tr = Transcript()
tr.chrom = "Chr1"
tr.start = 101
tr.end = 2000
tr.strand = "-"
tr.add_exons([(101, 300), (1701, 2000)])
tr.add_exons([(101, 300), (1701, 2000)], features="CDS")
tr.id = "test1"
tr.parent = "gene1"
# Phase 0, 0 because the first CDS exon is 300bp
gff = tr.format("gff", with_introns=True)
self.maxDiff = None
res = """Chr1\tMikado\tmRNA\t101\t2000\t.\t-\t.\tID=test1;Parent=gene1;Name=test1
Chr1\tMikado\tCDS\t101\t300\t.\t-\t0\tID=test1.CDS1;Parent=test1
Chr1\tMikado\texon\t101\t300\t.\t-\t.\tID=test1.exon1;Parent=test1
Chr1\tMikado\tintron\t301\t1700\t.\t-\t.\tID=test1.intron1;Parent=test1
Chr1\tMikado\tCDS\t1701\t2000\t.\t-\t0\tID=test1.CDS2;Parent=test1
Chr1\tMikado\texon\t1701\t2000\t.\t-\t.\tID=test1.exon2;Parent=test1"""
self.assertEqual(
gff, res, "++++\n\n" + "\n+++\n".join(
[gff, res, ",\t".join([str(_) for _ in tr.internal_orfs])]))
gtf = tr.format("gtf", with_introns=True)
res = """Chr1\tMikado\tmRNA\t101\t2000\t.\t-\t.\tgene_id "gene1"; transcript_id "test1"; Name "test1";
Chr1\tMikado\tCDS\t101\t300\t.\t-\t0\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t101\t300\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\tintron\t301\t1700\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\tCDS\t1701\t2000\t.\t-\t0\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t1701\t2000\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";"""
self.assertEqual(gtf, res, "++++\n\n" + "\n+++\n".join([gtf, res]))
def test_non_coding_negative(self):
tr = Transcript()
tr.chrom = "Chr1"
tr.start = 101
tr.end = 2000
tr.strand = "-"
tr.add_exons([(101, 300),
(1701, 2000)])
tr.id = "test1"
tr.parent = "gene1"
tr.finalize()
gff = tr.format("gff", with_introns=True)
self.maxDiff = None
res = """Chr1\tMikado\ttranscript\t101\t2000\t.\t-\t.\tID=test1;Parent=gene1
Chr1\tMikado\texon\t101\t300\t.\t-\t.\tID=test1.exon1;Parent=test1
Chr1\tMikado\tintron\t301\t1700\t.\t-\t.\tID=test1.intron1;Parent=test1
Chr1\tMikado\texon\t1701\t2000\t.\t-\t.\tID=test1.exon2;Parent=test1"""
self.assertEqual(gff,
res,
"++++\n\n"+"\n+++\n".join([gff, res]))
gtf = tr.format("gtf", with_introns=True)
res = """Chr1\tMikado\ttranscript\t101\t2000\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t101\t300\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\tintron\t301\t1700\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";
Chr1\tMikado\texon\t1701\t2000\t.\t-\t.\tgene_id "gene1"; transcript_id "test1";"""
self.assertEqual(gtf, res,
"++++\n\n" + "\n+++\n".join([gtf, res]))
def test_non_redundant_as(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 20
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.locus.add_transcript_to_locus(t2)
self.assertEqual(len(self.locus.transcripts), 2,
self.locus.transcripts)
t3 = Transcript()
t3.chrom = "Chr1"
t3.strand = "+"
t3.score = 20
t3.id = "G3.1"
t3.parent = "G3"
t3.start = 201
t3.end = 1630
t3.add_exons([(201, 500), (601, 670), (1031, 1300), (1401, 1460),
(1501, 1630)], "exon")
t3.add_exons([(401, 500), (601, 670), (1031, 1300), (1401, 1440)],
"CDS")
t3.logger = self.logger
t3.finalize()
self.assertEqual(
self.locus.is_alternative_splicing(t3)[:2], (True, "j"))
self.locus.add_transcript_to_locus(t3)
self.assertEqual(len(self.locus.transcripts), 3,
self.locus.transcripts)
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_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_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_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))
def test_not_intersecting(self):
# This one is contained and should be rejected
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 20
t2.id = "G1.1"
t2.parent = "G1"
t2.start = 601
t2.end = 1420
t2.add_exons([(601, 700), (1001, 1300), (1401, 1420)], "exon")
t2.add_exons([(601, 700), (1001, 1300), (1401, 1420)], "CDS")
t2.finalize()
self.assertEqual(
self.locus.is_alternative_splicing(t2)[:2], (False, "c"))
def test_creation_from_transcript(self):
t = Transcript()
t.chrom = "Chr1"
t.start = 100
t.end = 200
t.strand = "+"
t.id = "test"
t.parent = "parent"
gene = Gene(t)
self.assertIn(t.id, gene, gene.keys())
self.assertIn(t, gene, gene.keys())
self.assertEqual(t.chrom, gene.chrom)
self.assertEqual(t.start, gene.start)
self.assertEqual(t.end, gene.end)
self.assertEqual(t.strand, gene.strand)
self.assertIs(t, gene[t.id])
def test_creation_from_transcript(self):
t = Transcript()
t.chrom = "Chr1"
t.start = 100
t.end = 200
t.strand = "+"
t.id = "test"
t.parent = "parent"
gene = Gene(t)
self.assertIn(t.id, gene, gene.keys())
self.assertIn(t, gene, gene.keys())
self.assertEqual(t.chrom, gene.chrom)
self.assertEqual(t.start, gene.start)
self.assertEqual(t.end, gene.end)
self.assertEqual(t.strand, gene.strand)
self.assertIs(t, gene[t.id])
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_lowscore(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.locus.add_transcript_to_locus(t2)
self.assertEqual(len(self.locus.transcripts), 2,
self.locus.transcripts)
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_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_valid_as(self):
t2 = Transcript()
t2.chrom = "Chr1"
t2.strand = "+"
t2.score = 20
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.assertEqual(
self.locus.is_alternative_splicing(t2)[:2], (True, "J"))
self.locus.add_transcript_to_locus(t2)
self.assertEqual(len(self.locus.transcripts), 2,
self.locus.transcripts)
def test_correct_cds(self):
transcript = Transcript()
transcript.chrom = "Chr1"
transcript.source = "test"
transcript.start = 10000
transcript.end = 20000
exons = [(10000, 11500), (12000, 13000), (15000, 18000),
(19000, 20000)]
cds = [
(11400, 11500), # 101
(12000, 13000), # 1001 ==> 1102
(15000, 17998)
] # 2998 == > 3090 (y)
transcript.add_exons(exons)
transcript.add_exons(cds, features="CDS")
transcript.strand = "+"
transcript.finalize()
logger = Mikado.utilities.log_utils.create_null_logger("correct_cds")
copied = transcript.deepcopy()
trimmed = trim_coding(copied, logger, max_length=50)
self.assertEqual(trimmed.start, 11400)
self.assertEqual(trimmed.end, 19050)
copied = transcript.deepcopy()
self.assertEqual(copied.start, 10000)
trimmed = trim_coding(copied, logger, max_length=200)
self.assertEqual(trimmed.start, 11300)
self.assertEqual(trimmed.end, 19200)