def test_simple_split(self):
"""
Check that the original transcript can be split in two
:return:
"""
self.assertIsNotNone(self.transcript.configuration)
self.transcript.configuration.pick.chimera_split.blast_check = False
# with self.assertLogs("test_mono", level="DEBUG") as log_split:
new_transcripts = list(splitting.split_by_cds(self.transcript))
self.assertEqual(len(new_transcripts), 2,
"\n".join(str(_) for _ in new_transcripts))
self.assertEqual(new_transcripts[0].start, self.transcript.start)
self.assertEqual(new_transcripts[1].end, self.transcript.end)
sl = loci.Superlocus(self.transcript,
configuration=self.transcript.configuration)
self.assertFalse(sl.configuration.pick.chimera_split.blast_check)
self.assertEqual(len(sl.transcripts), 1)
sl.logger.setLevel("DEBUG")
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), 2)
def test_spanning_hit_lenient(self):
self.transcript.blast_hits = [self.get_spanning_hit()]
self.transcript.json_conf["pick"]["chimera_split"]["blast_check"] = True
self.transcript.json_conf["pick"][
"chimera_split"]["blast_params"]["leniency"] = "LENIENT"
self.assertEqual(1,
len(list(splitting.split_by_cds(self.transcript))))
def test_lenient_split_twohits(self):
hit2 = self.get_second_hit()
self.transcript.blast_hits.append(hit2)
self.transcript.json_conf["pick"]["chimera_split"]["blast_check"] = True
self.transcript.json_conf["pick"][
"chimera_split"]["blast_params"]["leniency"] = "LENIENT"
self.assertEqual(2,
len(list(splitting.split_by_cds(self.transcript))))
def testPositive(self):
self.bed1 = parsers.bed12.BED12()
self.header = False
self.bed1.chrom = "transcript1"
self.bed1.start = 1
self.bed1.end = 5000
self.bed1.name = "Bed1"
self.bed1.score = 0
self.bed1.strand = "+"
self.bed1.thick_start = 1
self.bed1.thick_end = 3002
self.bed1.phase = 2
self.bed1.block_counts = 1
self.bed1.block_sizes = [3002]
self.bed1.block_starts = [1]
self.bed1.transcriptomic = True
self.bed1.has_start_codon = False
self.bed1.has_stop_codon = True
self.assertFalse(self.bed1.invalid, self.bed1.invalid_reason)
self.bed2 = parsers.bed12.BED12()
self.header = False
self.bed2.chrom = "transcript1"
self.bed2.start = 1
self.bed2.end = 5000
self.bed2.name = "Bed2"
self.bed2.score = 0
self.bed2.strand = "+"
self.bed2.thick_start = 4001
self.bed2.thick_end = 4900
self.bed2.block_counts = 1
self.bed2.block_sizes = [900]
self.bed2.block_starts = [4001]
self.bed2.transcriptomic = True
self.bed2.has_start_codon = True
self.bed2.has_stop_codon = True
self.assertFalse(self.bed2.invalid)
self.transcript.load_orfs([self.bed1, self.bed2])
self.assertTrue(self.transcript.is_coding)
self.assertEqual(self.transcript.number_internal_orfs, 2,
str(self.transcript))
self.transcript.finalize()
# with self.assertLogs("test_mono", level="DEBUG") as log_split:
new_transcripts = list(splitting.split_by_cds(self.transcript))
# self.assertIn("DEBUG:test_mono:",
# log_split.output)
self.assertEqual(len(new_transcripts), 2,
"\n".join(str(_) for _ in new_transcripts))
self.assertEqual(new_transcripts[0].start, self.transcript.start)
self.assertEqual(new_transcripts[1].end, self.transcript.end)
def test_spanning_hit_nocheck(self):
self.transcript.blast_hits = [self.get_spanning_hit()]
self.transcript.json_conf["pick"]["chimera_split"]["blast_check"] = False
cds_boundaries = SortedDict()
for orf in sorted(self.transcript.loaded_bed12,
key=operator.attrgetter("thick_start", "thick_end")):
cds_boundaries[(orf.thick_start, orf.thick_end)] = [orf]
self.assertEqual(len(cds_boundaries), 2)
self.assertEqual(self.transcript.number_internal_orfs, 2)
self.assertEqual(2,
len(list(splitting.split_by_cds(self.transcript))))
def test_deleted_hits(self):
delattr(self.transcript, "blast_hits")
self.transcript.json_conf["pick"]["chimera_split"]["blast_check"] = True
self.transcript.json_conf["pick"][
"chimera_split"]["blast_params"]["leniency"] = "LENIENT"
self.transcript.logger = self.logger
with self.assertLogs("null", level="WARNING") as log_split:
self.assertEqual(2,
len(list(splitting.split_by_cds(self.transcript))))
self.assertIn("WARNING:null:BLAST hits store lost for transcript1! Creating a mock one to avoid a crash",
log_split.output)
def test_deleted_hits(self):
delattr(self.transcript, "blast_hits")
self.transcript.configuration.pick.chimera_split.blast_check = True
self.transcript.configuration.pick.chimera_split.blast_params.leniency = "LENIENT"
self.transcript.logger = self.logger
with self.assertLogs(logger=self.logger, level="WARNING") as log_split:
self.assertEqual(
2, len(list(splitting.split_by_cds(self.transcript))))
self.assertIn(
"WARNING:test_mono:BLAST hits store lost for transcript1! Creating a mock one to avoid a crash",
log_split.output)
def test_spanning_hit_lenient(self):
self.transcript.blast_hits = [self.get_spanning_hit()]
self.transcript.configuration.pick.chimera_split.blast_check = True
self.transcript.configuration.pick.chimera_split.blast_params.leniency = "LENIENT"
self.assertEqual(1, len(list(splitting.split_by_cds(self.transcript))))
sl = loci.Superlocus(self.transcript,
configuration=self.transcript.configuration)
self.assertEqual(len(sl.transcripts), 1)
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), 1)
def test_lenient_split_twohits(self):
hit2 = self.get_second_hit()
self.transcript.blast_hits.append(hit2)
self.transcript.configuration.pick.chimera_split.blast_check = True
self.transcript.configuration.pick.chimera_split.blast_params.leniency = "LENIENT"
self.assertEqual(2, len(list(splitting.split_by_cds(self.transcript))))
sl = loci.Superlocus(self.transcript,
configuration=self.transcript.configuration)
self.assertEqual(len(sl.transcripts), 1)
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), 2)
def test_no_hsps(self):
self.transcript.blast_hits = []
self.transcript.configuration.pick.chimera_split.blast_check = True
self.transcript.configuration.pick.chimera_split.blast_params.leniency = "LENIENT"
logger = utilities.log_utils.create_default_logger("test_no_hsps")
logger.setLevel("DEBUG")
self.transcript.logger = logger
self.assertEqual(2, len(list(splitting.split_by_cds(self.transcript))))
sl = loci.Superlocus(self.transcript,
configuration=self.transcript.configuration)
self.assertEqual(len(sl.transcripts), 1)
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), 2)
def test_spanning_hit_nocheck(self):
self.transcript.blast_hits = [self.get_spanning_hit()]
self.transcript.configuration.pick.chimera_split.blast_check = False
cds_boundaries = SortedDict()
for orf in sorted(self.transcript.loaded_bed12,
key=operator.attrgetter("thick_start", "thick_end")):
cds_boundaries[(orf.thick_start, orf.thick_end)] = [orf]
self.assertEqual(len(cds_boundaries), 2)
self.assertEqual(self.transcript.number_internal_orfs, 2)
self.assertEqual(2, len(list(splitting.split_by_cds(self.transcript))))
sl = loci.Superlocus(self.transcript,
configuration=self.transcript.configuration)
self.assertEqual(len(sl.transcripts), 1)
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), 2)
def test_no_splitting_by_source(self):
self.transcript.source = "foo"
for sources in [[], [self.transcript.source], ["bar"],
["bar", [self.transcript.source]]]:
with self.subTest(sources=sources):
self.transcript.configuration.pick.chimera_split.skip = sources
if self.transcript.source in sources:
final = 1
else:
final = 2
self.assertEqual(
final, len(list(splitting.split_by_cds(self.transcript))))
sl = loci.Superlocus(
self.transcript,
configuration=self.transcript.configuration)
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), final)
def test_simple_split(self):
"""
Check that the original transcript can be split in two
:return:
"""
self.assertIsNotNone(self.transcript.json_conf)
self.assertIn("pick", self.transcript.json_conf)
self.transcript.json_conf["pick"]["chimera_split"]["blast_check"] = False
# with self.assertLogs("test_mono", level="DEBUG") as log_split:
new_transcripts = list(splitting.split_by_cds(self.transcript))
# print(log_split.output)
# self.assertIn("DEBUG:test_mono:",
# log_split.output)
self.assertEqual(len(new_transcripts), 2, "\n".join(str(_) for _ in new_transcripts))
self.assertEqual(new_transcripts[0].start, self.transcript.start)
self.assertEqual(new_transcripts[1].end, self.transcript.end)
def testNegative(self):
self.bed1 = parsers.bed12.BED12()
self.header = False
self.bed1.chrom = "transcript1"
self.bed1.start = 1
self.bed1.end = 5000
self.bed1.name = "Bed1"
self.bed1.score = 0
self.bed1.strand = "-"
self.bed1.thick_start = 1
self.bed1.thick_end = 3000
self.bed1.phase = 0
self.bed1.block_counts = 1
self.bed1.block_sizes = [3001]
self.bed1.block_starts = [1]
self.bed1.transcriptomic = True
self.bed1.has_start_codon = True
self.bed1.has_stop_codon = True
self.assertFalse(self.bed1.invalid, self.bed1.invalid_reason)
logger = create_default_logger("testNegative", "DEBUG")
self.bed2 = parsers.bed12.BED12(logger=logger)
self.header = False
self.bed2.chrom = "transcript1"
self.bed2.start = 1
self.bed2.end = 5000
self.bed2.name = "Bed2"
self.bed2.score = 0
self.bed2.strand = "-"
self.bed2.thick_start = 4001
self.bed2.thick_end = 5000
self.bed2.phase = 1
self.bed2.block_counts = 1
self.bed2.block_sizes = [1000]
self.bed2.block_starts = [4001]
self.bed2.transcriptomic = True
self.bed2.has_start_codon = False
self.bed2.has_stop_codon = True
self.assertFalse(self.bed2.invalid,
(self.bed2.phase, self.bed2.invalid_reason))
self.transcript.load_orfs([self.bed1, self.bed2])
self.assertTrue(self.transcript.is_coding)
self.assertEqual(self.transcript.number_internal_orfs, 2,
str(self.transcript))
self.transcript.finalize()
self.assertEqual(self.transcript.number_internal_orfs, 2,
str(self.transcript))
# with self.assertLogs("test_mono", level="DEBUG") as log_split:
new_transcripts = list(splitting.split_by_cds(self.transcript))
self.assertEqual(len(new_transcripts), 2,
"\n".join(str(_) for _ in new_transcripts))
self.assertEqual(new_transcripts[0].start, self.transcript.start)
self.assertEqual(new_transcripts[1].end, self.transcript.end)
self.assertEqual(self.transcript.combined_cds_start, 6000)
self.assertEqual(self.transcript.combined_cds_end, 1001)
self.assertEqual(self.transcript.selected_cds_end, 1001)
self.assertEqual(self.transcript.selected_cds_start, 4000)
self.assertEqual(self.transcript.strand, "-")
sl = loci.Superlocus(self.transcript,
configuration=self.transcript.configuration)
self.assertEqual(len(sl.transcripts), 1)
sl.load_all_transcript_data()
self.assertEqual(len(sl.transcripts), 2)
def test_one_orf(self):
self.transcript.strip_cds()
self.transcript.load_orfs([self.bed1])
self.assertEqual(1, len(list(splitting.split_by_cds(self.transcript))))