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_locus(self):
"""Basic testing of the Locus functionality."""
logger = create_null_logger("null")
logger.setLevel("WARNING")
logger.info("Started")
slocus = Superlocus(self.transcript1,
json_conf=self.my_json,
logger=logger)
slocus.add_transcript_to_locus(self.transcript2)
self.assertEqual(slocus.strand, self.transcript1.strand)
self.assertEqual(slocus.start,
min(self.transcript1.start, self.transcript2.start))
self.assertEqual(slocus.end,
max(self.transcript1.end, self.transcript2.end))
logger.info(slocus.transcripts)
slocus.define_subloci()
logger.info(slocus.subloci)
logger.info(slocus.transcripts)
self.assertEqual(len(slocus.transcripts), 2)
self.assertEqual(len(slocus.subloci), 2)
slocus.define_monosubloci()
self.assertEqual(len(slocus.monosubloci), 2)
slocus.define_loci()
self.assertEqual(len(slocus.loci), 1)
self.assertEqual(
list(slocus.loci[list(
slocus.loci.keys())[0]].transcripts.keys())[0], "t0")
gff_transcript3 = """Chr1\tfoo\ttranscript\t101\t200\t.\t-\t.\tID=tminus0
Chr1\tfoo\texon\t101\t200\t.\t-\t.\tID=tminus0:exon1;Parent=tminus0""".split(
"\n")
gff_transcript3 = [GFF.GffLine(x) for x in gff_transcript3]
transcript3 = Transcript(gff_transcript3[0])
for exon in gff_transcript3[1:]:
transcript3.add_exon(exon)
transcript3.finalize()
minusuperlocus = Superlocus(transcript3, json_conf=self.my_json)
minusuperlocus.define_loci()
self.assertEqual(len(minusuperlocus.loci), 1)
self.assertTrue(transcript3.strand != self.transcript1.strand)
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)
def main():
parser = argparse.ArgumentParser(__doc__)
parser.add_argument("-mi",
"--max-intron",
default=10000,
dest="max_intron",
type=int,
help="Maximum intron length for UTR introns.")
parser.add_argument("gff", type=parser_factory)
parser.add_argument("out",
default=sys.stdout,
type=argparse.FileType("wt"),
nargs="?")
args = parser.parse_args()
if args.max_intron < 0:
raise ValueError("Max intron length <0 specified! {0}".format(
args.max_intron))
ref_gff = isinstance(args.gff, GFF3)
if ref_gff:
form = "gff3"
else:
form = "gtf"
current = None
current_transcript = None
last_header = []
for record in args.gff:
if record.header is True:
# print(record, file=sys.stderr)
if current is not None:
current = remove_introns(current, args)
print(current.format(form), file=args.out)
print(*last_header, sep="\n", file=args.out, end='')
current = None
current_transcript = None
print(*last_header, sep="\n", end="")
last_header = [record]
continue
if record.feature not in ("gene", "mRNA", "CDS", "exon"):
continue
if record.is_gene is True and ref_gff:
print(record, file=sys.stderr)
last_header = []
if current is not None:
# current = remove_introns(current, args)
print(current.format(form), file=args.out)
print(*last_header, sep="\n", file=args.out, end='')
current = None
current_transcript = None
if record.is_transcript:
if ref_gff is False:
if current_transcript is not None:
# current_transcript = remove_introns_from_transcr(current_transcript,
# args)
assert current_transcript.combined_cds_length > 0
print(current_transcript, file=args.out)
print(*last_header, sep="\n", file=args.out)
last_header = []
elif ref_gff is True:
if current_transcript is not None:
if current is None:
current = Gene(current_transcript)
current.add(current_transcript)
else:
assert current_transcript.parent[0] != current.id
current.add(current_transcript)
# if current.id == current_transcript.parent[0]:
# else:
# current = remove_introns(current, args)
# print(current.format(form), file=args.out)
# print("###", file=args.out)
# current = None
# elif current_transcript is not None:
# current = Gene(current_transcript)
current_transcript = Transcript(record)
elif record.is_exon:
if record.feature not in ("CDS", "exon"):
continue
current_transcript.add_exon(record)
else:
continue
continue
if ref_gff and current is not None:
print(*last_header, sep="\n", file=args.out)
last_header = []
current = remove_introns(current, args)
print(current.format(form), file=args.out)
elif not ref_gff and current_transcript is not None:
current_transcript = remove_introns_from_transcr(
current_transcript, args)
print(current_transcript.format(form), file=args.out)
print(*last_header, sep="\n", file=args.out, end='')
def main():
parser = argparse.ArgumentParser(__doc__)
parser.add_argument("-mi", "--max-intron", default=10000, dest="max_intron",
type=int, help="Maximum intron length for UTR introns.")
parser.add_argument("gff", type=to_gff)
parser.add_argument("out", default=sys.stdout, type=argparse.FileType("wt"), nargs="?")
args = parser.parse_args()
if args.max_intron < 0:
raise ValueError("Max intron length <0 specified! {0}".format(args.max_intron))
ref_gff = isinstance(args.gff, GFF3)
if ref_gff:
form = "gff3"
else:
form = "gtf"
current = None
current_transcript = None
last_header = []
for record in args.gff:
if record.header is True:
# print(record, file=sys.stderr)
if current is not None:
current = remove_introns(current, args)
print(current.format(form), file=args.out)
print(*last_header, sep="\n", file=args.out, end='')
current = None
current_transcript = None
print(*last_header, sep="\n", end="")
last_header = [record]
continue
if record.feature not in ("gene", "mRNA", "CDS", "exon"):
continue
if record.is_gene is True and ref_gff:
print(record, file=sys.stderr)
last_header = []
if current is not None:
# current = remove_introns(current, args)
print(current.format(form), file=args.out)
print(*last_header, sep="\n", file=args.out, end='')
current = None
current_transcript = None
if record.is_transcript:
if ref_gff is False:
if current_transcript is not None:
# current_transcript = remove_introns_from_transcr(current_transcript,
# args)
assert current_transcript.combined_cds_length > 0
print(current_transcript, file=args.out)
print(*last_header, sep="\n", file=args.out)
last_header = []
elif ref_gff is True:
if current_transcript is not None:
if current is None:
current = Gene(current_transcript)
current.add(current_transcript)
else:
assert current_transcript.parent[0] != current.id
current.add(current_transcript)
# if current.id == current_transcript.parent[0]:
# else:
# current = remove_introns(current, args)
# print(current.format(form), file=args.out)
# print("###", file=args.out)
# current = None
# elif current_transcript is not None:
# current = Gene(current_transcript)
current_transcript = Transcript(record)
elif record.is_exon:
if record.feature not in ("CDS", "exon"):
continue
current_transcript.add_exon(record)
else:
continue
continue
if ref_gff and current is not None:
print(*last_header, sep="\n", file=args.out)
last_header = []
current = remove_introns(current, args)
print(current.format(form), file=args.out)
elif not ref_gff and current_transcript is not None:
current_transcript = remove_introns_from_transcr(current_transcript,
args)
print(current_transcript.format(form), file=args.out)
print(*last_header, sep="\n", file=args.out, end='')
class LocusTester(unittest.TestCase):
logger = create_null_logger("locus_tester")
def setUp(self):
gff_transcript1 = """Chr1\tfoo\ttranscript\t101\t300\t.\t+\t.\tID=t0
Chr1\tfoo\texon\t101\t300\t.\t+\t.\tID=t0:exon1;Parent=t0
Chr1\tfoo\tCDS\t101\t250\t.\t+\t.\tID=t0:exon1;Parent=t0""".split("\n")
gff_transcript1 = [GFF.GffLine(x) for x in gff_transcript1]
self.assertEqual(gff_transcript1[0].chrom, "Chr1", gff_transcript1[0])
self.transcript1 = Transcript(gff_transcript1[0])
for exon in gff_transcript1[1:]:
self.transcript1.add_exon(exon)
self.transcript1.finalize()
self.assertTrue(self.transcript1.monoexonic)
self.assertEqual(self.transcript1.chrom, gff_transcript1[0].chrom)
gff_transcript2 = """Chr1\tfoo\ttranscript\t101\t600\t.\t+\t.\tID=t1
Chr1\tfoo\texon\t101\t200\t.\t+\t.\tID=t1:exon1;Parent=t1
Chr1\tfoo\texon\t301\t400\t.\t+\t.\tID=t1:exon2;Parent=t1
Chr1\tfoo\texon\t501\t600\t.\t+\t.\tID=t1:exon3;Parent=t1""".split("\n")
gff_transcript2 = [GFF.GffLine(x) for x in gff_transcript2]
self.transcript2 = Transcript(gff_transcript2[0], logger=self.logger)
for exon in gff_transcript2[1:-1]:
self.transcript2.add_exon(exon)
# Test that a transcript cannot be finalized if
# the exons do not define the external boundaries
with self.assertLogs("null", level="WARNING") as _:
self.transcript2.finalize()
with self.assertRaises(exceptions.ModificationError):
self.transcript2.add_exon(gff_transcript2[-1])
self.transcript2.finalized = False
self.transcript2.start = 101
self.transcript2.end = 600
self.transcript2.add_exon(gff_transcript2[-1])
self.transcript2.finalize()
self.assertFalse(self.transcript2.monoexonic)
self.assertEqual(self.transcript2.exon_num, len(gff_transcript2) - 1)
# Test that trying to modify a transcript after it has been finalized causes errors
with self.assertRaises(exceptions.ModificationError):
for exon in gff_transcript2[1:]:
self.transcript2.add_exon(exon)
# Test that creating a superlocus without configuration fails
with self.assertRaises(exceptions.NoJsonConfigError):
_ = Superlocus(self.transcript1)
self.my_json = os.path.join(os.path.dirname(__file__),
"configuration.yaml")
self.my_json = configurator.to_json(self.my_json)
self.assertIn("scoring", self.my_json, self.my_json.keys())
def test_locus(self):
"""Basic testing of the Locus functionality."""
logger = create_null_logger("null")
logger.setLevel("WARNING")
logger.info("Started")
slocus = Superlocus(self.transcript1,
json_conf=self.my_json,
logger=logger)
slocus.add_transcript_to_locus(self.transcript2)
self.assertEqual(slocus.strand, self.transcript1.strand)
self.assertEqual(slocus.start,
min(self.transcript1.start, self.transcript2.start))
self.assertEqual(slocus.end,
max(self.transcript1.end, self.transcript2.end))
logger.info(slocus.transcripts)
slocus.define_subloci()
logger.info(slocus.subloci)
logger.info(slocus.transcripts)
self.assertEqual(len(slocus.transcripts), 2)
self.assertEqual(len(slocus.subloci), 2)
slocus.define_monosubloci()
self.assertEqual(len(slocus.monosubloci), 2)
slocus.define_loci()
self.assertEqual(len(slocus.loci), 1)
self.assertEqual(
list(slocus.loci[list(
slocus.loci.keys())[0]].transcripts.keys())[0], "t0")
gff_transcript3 = """Chr1\tfoo\ttranscript\t101\t200\t.\t-\t.\tID=tminus0
Chr1\tfoo\texon\t101\t200\t.\t-\t.\tID=tminus0:exon1;Parent=tminus0""".split(
"\n")
gff_transcript3 = [GFF.GffLine(x) for x in gff_transcript3]
transcript3 = Transcript(gff_transcript3[0])
for exon in gff_transcript3[1:]:
transcript3.add_exon(exon)
transcript3.finalize()
minusuperlocus = Superlocus(transcript3, json_conf=self.my_json)
minusuperlocus.define_loci()
self.assertEqual(len(minusuperlocus.loci), 1)
self.assertTrue(transcript3.strand != self.transcript1.strand)
