def setUp(self) -> None:
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'gembase'
args.models_dir = self.find_data('models')
cfg = Config(MacsyDefaults(), args)
model_name = 'foo'
models_location = ModelLocation(path=os.path.join(args.models_dir, model_name))
model = Model("foo/T2SS", 10)
profile_factory = ProfileFactory(cfg)
gene_name = "gspD"
self.cg_gspd = CoreGene(models_location, gene_name, profile_factory)
self.mg_gspd = ModelGene(self.cg_gspd, model, loner=True, multi_system=True)
gene_name = "sctJ"
self.cg_sctj = CoreGene(models_location, gene_name, profile_factory)
self.mg_sctj = ModelGene(self.cg_sctj, model)
model.add_mandatory_gene(self.mg_gspd)
model.add_accessory_gene(self.mg_sctj)
self.chit_1 = CoreHit(self.cg_gspd, "hit_1", 803, "replicon_id", 2, 1.0, 1.0, 1.0, 1.0, 10, 20)
self.chit_2 = CoreHit(self.cg_sctj, "hit_2", 803, "replicon_id", 3, 1.0, 1.0, 1.0, 1.0, 10, 20)
self.chit_3 = CoreHit(self.cg_gspd, "hit_3", 803, "replicon_id", 10, 1.0, 1.0, 1.0, 1.0, 10, 20)
self.chit_4 = CoreHit(self.cg_gspd, "hit_4", 803, "replicon_id", 20, 1.0, 1.0, 1.0, 1.0, 10, 20)
self.mhit_1 = ModelHit(self.chit_1, self.mg_gspd, GeneStatus.MANDATORY)
self.mhit_2 = ModelHit(self.chit_2, self.mg_sctj, GeneStatus.ACCESSORY)
self.mhit_3 = ModelHit(self.chit_3, self.mg_gspd, GeneStatus.MANDATORY)
self.mhit_4 = ModelHit(self.chit_4, self.mg_gspd, GeneStatus.MANDATORY)
def test_str(self):
model = Model("foo/T2SS", 10)
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
c_gene_sctn = CoreGene(self.model_location, "sctN",
self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model)
model.add_accessory_gene(gene_sctn)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 2, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = CoreHit(c_gene_sctn, "hit_3", 803, "replicon_id", 3, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_3 = ModelHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
uls_1 = UnlikelySystem(model, [v_hit_1], [v_hit_2, v_hit_3], [], [],
["reason"])
expected_str = """(hit_1, gspD, 1), (hit_2, sctJ, 2), (hit_3, sctN, 3): These hits does not probably constitute a system because:
reason"""
self.assertEqual(str(uls_1), expected_str)
def test_get_position(self):
gene_name = "gspD"
gene = CoreGene(self.model_location, gene_name, self.profile_factory)
h0 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
self.assertEqual(h0.get_position(), 3450)
def test_init(self):
model = Model("foo/model_A", 10)
# test if id is well incremented
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
ls_1 = LikelySystem(model, [v_hit_1], [v_hit_2], [], [])
self.assertTrue(ls_1.id.startswith('replicon_id_model_A_'))
ls_2 = LikelySystem(model, [v_hit_1, v_hit_2], [], [], [])
# check if the id of the second likelysystem is well increased
self.assertEqual(int(ls_2.id.split('_')[-1]),
int(ls_1.id.split('_')[-1]) + 1)
def test_hits(self):
model = Model("foo/T2SS", 10)
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
c_gene_sctn = CoreGene(self.model_location, "sctN",
self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model)
model.add_accessory_gene(gene_sctn)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = CoreHit(c_gene_sctn, "hit_3", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_3 = ModelHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
ls_1 = LikelySystem(model, [v_hit_1], [v_hit_2, v_hit_3], [], [])
self.assertListEqual(ls_1.hits, [v_hit_1, v_hit_2, v_hit_3])
def test_str(self):
model = Model("foo/T2SS", 10)
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
c_gene_sctn = CoreGene(self.model_location, "sctN",
self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model)
model.add_accessory_gene(gene_sctn)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = CoreHit(c_gene_sctn, "hit_3", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_3 = ModelHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
ls_1 = LikelySystem(model, [v_hit_1], [v_hit_2, v_hit_3], [], [])
expected_str = ', '.join([
f"({h.id}, {h.gene.name}, {h.position})"
for h in (v_hit_1, v_hit_2, v_hit_3)
])
self.assertEqual(str(ls_1), expected_str)
def test_UnlikelySystemSerializer_txt(self):
model = Model("foo/FOO", 10)
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
c_gene_sctn = CoreGene(self.model_location, "sctN",
self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model)
model.add_accessory_gene(gene_sctn)
c_gene_abc = CoreGene(self.model_location, "abc", self.profile_factory)
gene_abc = ModelGene(c_gene_abc, model)
model.add_forbidden_gene(gene_abc)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 2, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = CoreHit(c_gene_sctn, "hit_3", 803, "replicon_id", 3, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_3 = ModelHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
hit_4 = CoreHit(c_gene_abc, "hit_4", 803, "replicon_id", 4, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_4 = ModelHit(hit_4, gene_abc, GeneStatus.FORBIDDEN)
ser = TxtUnikelySystemSerializer()
ls_1 = UnlikelySystem(model, [v_hit_1], [v_hit_2, v_hit_3], [],
[v_hit_4], ["the reason why"])
txt = ser.serialize(ls_1)
expected_txt = """This replicon probably not contains a system foo/FOO:
the reason why
system id = replicon_id_FOO_1
model = foo/FOO
replicon = replicon_id
hits = [('hit_1', 'gspD', 1), ('hit_2', 'sctJ', 2), ('hit_3', 'sctN', 3), ('hit_4', 'abc', 4)]
wholeness = 1.000
mandatory genes:
\t- gspD: 1 (gspD)
accessory genes:
\t- sctJ: 1 (sctJ)
\t- sctN: 1 (sctN)
neutral genes:
forbidden genes:
\t- abc: 1 (abc)
Use ordered replicon to have better prediction.
"""
self.assertEqual(txt, expected_txt)
def setUp(self) -> None:
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'gembase'
args.models_dir = self.find_data('models')
cfg = Config(MacsyDefaults(), args)
model_name = 'foo'
self.models_location = ModelLocation(path=os.path.join(args.models_dir, model_name))
# we need to reset the ProfileFactory
# because it's a like a singleton
# so other tests are influenced by ProfileFactory and it's configuration
# for instance search_genes get profile without hmmer_exe
profile_factory = ProfileFactory(cfg)
model = Model(model_name, 10)
self.profile_factory = ProfileFactory(cfg)
gene_name = "gspD"
c_gene_gspd = CoreGene(self.models_location, gene_name, self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model, multi_system=True)
gene_name = "sctJ"
c_gene_sctj = CoreGene(self.models_location, gene_name, self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model, multi_system=True)
gene_name = "sctN"
c_gene_sctn = CoreGene(self.models_location, gene_name, self.profile_factory)
gene_sctn = Exchangeable(c_gene_sctn, gene_sctj)
gene_sctj.add_exchangeable(gene_sctn)
model.add_mandatory_gene(gene_gspd)
model.add_accessory_gene(gene_sctj)
# CoreHit(gene, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
# pos score
chit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 2, 1.0, 1.0, 1.0, 1.0, 10, 20)
chit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 3, 1.0, 1.0, 1.0, 1.0, 10, 20)
chit_3 = CoreHit(c_gene_gspd, "hit_3", 803, "replicon_id", 10, 1.0, 3.0, 1.0, 1.0, 10, 20)
chit_4 = CoreHit(c_gene_sctn, "hit_4", 803, "replicon_id", 14, 1.0, 4.0, 1.0, 1.0, 10, 20)
chit_5 = CoreHit(c_gene_gspd, "hit_5", 803, "replicon_id", 20, 1.0, 2.0, 1.0, 1.0, 10, 20)
self.mhit_1 = ModelHit(chit_1, gene_gspd, GeneStatus.MANDATORY)
self.mhit_2 = ModelHit(chit_2, gene_sctj, GeneStatus.ACCESSORY)
self.mhit_3 = ModelHit(chit_3, gene_gspd, GeneStatus.MANDATORY)
self.mhit_4 = ModelHit(chit_4, gene_sctn, GeneStatus.ACCESSORY)
self.mhit_5 = ModelHit(chit_5, gene_gspd, GeneStatus.MANDATORY)
self.ms_1 = MultiSystem(chit_1, gene_ref=gene_gspd, gene_status=GeneStatus.MANDATORY)
self.ms_2 = MultiSystem(chit_2, gene_ref=gene_sctj, gene_status=GeneStatus.ACCESSORY)
self.ms_3 = MultiSystem(chit_3, gene_ref=gene_gspd, gene_status=GeneStatus.MANDATORY)
self.ms_4 = MultiSystem(chit_4, gene_ref=gene_sctn, gene_status=GeneStatus.ACCESSORY)
self.ms_5 = MultiSystem(chit_5, gene_ref=gene_gspd, gene_status=GeneStatus.MANDATORY)
def test_eq(self):
gene_name = "gspD"
gene = CoreGene(self.model_location, gene_name, self.profile_factory)
h0 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h1 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h2 = CoreHit(gene, "PSAE001c01_013980", 759, "PSAE001c01", 4146, float(3.7e-76), float(255.8),
float(1.000000), (736.0 - 105.0 + 1) / 759, 105, 736)
self.assertEqual(h0, h1)
self.assertNotEqual(h0, h2)
def test_hash(self):
gene_name = "gspD"
gene = CoreGene(self.model_location, gene_name, self.profile_factory)
h0 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h1 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h2 = CoreHit(gene, "PSAE001c01_006941", 803, "PSAE001c01", 3450, float(1.2e-234), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
self.assertTrue(isinstance(hash(h0), int))
self.assertEqual(hash(h0), hash(h1))
self.assertNotEqual(hash(h0), hash(h2))
def test_search_recover(self):
# first job searching using hmmsearch
gene_name = "abc"
c_gene_abc = CoreGene(self.model_location, gene_name,
self.profile_factory)
report = search_genes([c_gene_abc], self.cfg)
expected_hit = [
CoreHit(c_gene_abc, "ESCO030p01_000260", 706, "ESCO030p01", 26,
float(1.000e-200), float(660.800), float(1.000),
float(0.714), 160, 663)
]
# second job using recover
# disable hmmer to be sure that test use the recover inner function
self.cfg.hmmer = lambda: "hmmer_disable"
# and create a new dir for the second job
previous_job_path = self.cfg.working_dir()
self.cfg.previous_run = lambda: previous_job_path
self.cfg.out_dir = lambda: os.path.join(self.tmp_dir, 'job_2')
os.mkdir(self.cfg.out_dir())
# rerun with previous run
# but we have to reset the profile attached to the gene gene._profile._report
self.profile_factory = ProfileFactory(self.cfg)
c_gene_abc = CoreGene(self.model_location, gene_name,
self.profile_factory)
report = search_genes([c_gene_abc], self.cfg)
self.assertEqual(len(report), 1)
self.assertEqual(expected_hit[0], report[0].hits[0])
def test_cmp(self):
gene_name = "gspD"
gene = CoreGene(self.model_location, gene_name, self.profile_factory)
# compare hit with different id (comparison based on seq identifier)
h0 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234),
float(779.2), float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h1 = CoreHit(gene, "PSAE001c01_013980", 759, "PSAE001c01", 4146, float(3.7e-76),
float(255.8), float(1.000000), (736.0 - 105.0 + 1) / 759, 105, 736)
self.assertGreater(h1, h0)
self.assertLess(h0, h1)
# compare hit with different same id (comparison based on score)
# score = 779.2
h0 = CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234),
float(779.2), float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
# score = 255.8
h1 = CoreHit(gene, "PSAE001c01_006940", 759, "PSAE001c01", 4146, float(3.7e-76),
float(255.8), float(1.000000), (736.0 - 105.0 + 1) / 759, 105, 736)
self.assertGreater(h0, h1)
self.assertLess(h1, h0)
def test_search(self):
gene_name = "abc"
c_gene_abc = CoreGene(self.model_location, gene_name,
self.profile_factory)
report = search_genes([c_gene_abc], self.cfg)
expected_hit = [
CoreHit(c_gene_abc, "ESCO030p01_000260", 706, "ESCO030p01", 26,
float(1.000e-200), float(660.800), float(1.000),
float(0.714), 160, 663)
]
self.assertEqual(len(report), 1)
self.assertEqual(expected_hit[0], report[0].hits[0])
def test_reason(self):
model = Model("foo/model_A", 10)
# test if id is well incremented
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_forbidden_gene(gene_sctj)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.FORBIDDEN)
reason_2 = ["forbidden gene"]
uls_2 = UnlikelySystem(model, [v_hit_1], [], [], [v_hit_2], reason_2)
self.assertEqual(uls_2.reasons, reason_2)
def test_extract_concurent(self):
gene_name = "gspD"
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
shutil.copy(
self.find_data("hmm", gene_name + self.cfg.res_search_suffix()),
self.cfg.working_dir())
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
reports = []
for i in range(5):
report = OrderedHMMReport(c_gene, report_path, self.cfg)
reports.append(report)
import threading
def worker(report):
report.extract()
for report in reports:
t = threading.Thread(target=worker, args=(report, ))
t.start()
main_thread = threading.currentThread()
for t in threading.enumerate():
if t is main_thread:
continue
t.join()
# gene, model, hit_id, hit_seq_length replicon_name, pos_hit, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
hits = [
CoreHit(c_gene, "NC_xxxxx_xx_056141", 803,
RepliconDB.ordered_replicon_name, 141, float(2e-236),
float(779.2), float(1.000000), (741.0 - 104.0 + 1) / 803,
104, 741),
CoreHit(c_gene, "PSAE001c01_006940", 803,
RepliconDB.ordered_replicon_name, 68, float(1.2e-234),
float(779.2), float(1.000000), (741.0 - 104.0 + 1) / 803,
104, 741),
CoreHit(c_gene, "PSAE001c01_013980", 759,
RepliconDB.ordered_replicon_name, 69, float(3.7e-76),
float(255.8), float(1.000000), (736.0 - 105.0 + 1) / 759,
105, 736),
CoreHit(c_gene, "PSAE001c01_017350", 600,
RepliconDB.ordered_replicon_name, 70, float(3.2e-27),
float(94.2), float(0.500000), (506.0 - 226.0 + 1) / 600,
226, 506),
CoreHit(c_gene, "PSAE001c01_018920", 776,
RepliconDB.ordered_replicon_name, 71, float(6.1e-183),
float(608.4), float(1.000000), (606.0 - 48.0 + 1) / 776,
48, 606),
CoreHit(c_gene, "PSAE001c01_031420", 658,
RepliconDB.ordered_replicon_name, 73, float(1.8e-210),
float(699.3), float(1.000000), (614.0 - 55.0 + 1) / 658,
55, 614)
]
for report in reports:
report.save_extract()
self.assertEqual(len(report.hits), len(hits))
self.assertListEqual(report.hits, hits)
def test_extract(self):
gene_name = "gspD"
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
shutil.copy(
self.find_data("hmm", gene_name + self.cfg.res_search_suffix()),
self.cfg.working_dir())
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
report = OrderedHMMReport(c_gene, report_path, self.cfg)
report.extract()
self.assertEqual(len(report.hits), 6)
# gene, model, hit_id, hit_seq_ length replicon_name, pos_hit, i_eval,
# score, profile_coverage, sequence_coverage, begin_match, end_match
hits = [
CoreHit(c_gene, "NC_xxxxx_xx_056141", 803,
RepliconDB.ordered_replicon_name, 141, float(2e-236),
float(779.2), float(1.000000), (741.0 - 104.0 + 1) / 803,
104, 741),
CoreHit(c_gene, "PSAE001c01_006940", 803,
RepliconDB.ordered_replicon_name, 68, float(1.2e-234),
float(779.2), float(1.000000), (741.0 - 104.0 + 1) / 803,
104, 741),
CoreHit(c_gene, "PSAE001c01_013980", 759,
RepliconDB.ordered_replicon_name, 69, float(3.7e-76),
float(255.8), float(1.000000), (736.0 - 105.0 + 1) / 759,
105, 736),
CoreHit(c_gene, "PSAE001c01_017350", 600,
RepliconDB.ordered_replicon_name, 70, float(3.2e-27),
float(94.2), float(0.500000), (506.0 - 226.0 + 1) / 600,
226, 506),
CoreHit(c_gene, "PSAE001c01_018920", 776,
RepliconDB.ordered_replicon_name, 71, float(6.1e-183),
float(608.4), float(1.000000), (606.0 - 48.0 + 1) / 776,
48, 606),
CoreHit(c_gene, "PSAE001c01_031420", 658,
RepliconDB.ordered_replicon_name, 73, float(1.8e-210),
float(699.3), float(1.000000), (614.0 - 55.0 + 1) / 658,
55, 614)
]
self.assertListEqual(hits, report.hits)
report = OrderedHMMReport(c_gene, report_path, self.cfg)
report.hits = hits
self.assertIsNone(report.extract())
index_file = self.cfg.sequence_db() + '.idx'
with open(index_file, 'r') as idx_file:
idx = idx_file.readlines()
idx = idx[:-1]
with open(index_file, 'w') as idx_file:
idx_file.writelines(idx)
report = OrderedHMMReport(c_gene, report_path, self.cfg)
with self.assertRaises(MacsypyError) as ctx:
with self.catch_log() as log:
report.extract()
self.assertEqual(
str(ctx.exception),
"hit id 'NC_xxxxx_xx_056141' was not indexed, rebuild sequence 'test_base.fa' index"
)
def test_get_best_hits_4_func(self):
model = Model("foo/T2SS", 10)
gene_name = "gspD"
c_gene_gspd = CoreGene(self.models_location, gene_name, self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model, loner=True)
# gene, model, id, hit_seq_len, replicon_name, position, i_eval,
# score, profil_coverage, sequence_coverage, begin,end
######################
# based on the score #
######################
h0 = CoreHit(gene_gspd, "PSAE001c01_006940", 803, "PSAE001c01", 3450, float(1.2e-234),
10, float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h1 = CoreHit(gene_gspd, "PSAE001c01_013980", 759, "PSAE001c01", 3450, float(3.7e-76),
11, float(1.000000), (736.0 - 105.0 + 1) / 759, 105, 736)
m0 = ModelHit(h0, gene_gspd, GeneStatus.ACCESSORY)
m1 = ModelHit(h1, gene_gspd, GeneStatus.ACCESSORY)
l0 = Loner(h0, gene_ref=gene_gspd, gene_status=GeneStatus.ACCESSORY, counterpart=[m1])
l1 = Loner(h1, gene_ref=gene_gspd, gene_status=GeneStatus.ACCESSORY, counterpart=[m0])
l = get_best_hit_4_func(gene_name, [l0, l1])
self.assertEqual(l, l1)
#######################
# based on the i_eval #
#######################
h0 = CoreHit(gene_gspd, "PSAE001c01_006940", 803, "PSAE001c01", 3450, 10,
10, float(1.000000), (741.0 - 104.0 + 1) / 803, 104, 741)
h1 = CoreHit(gene_gspd, "PSAE001c01_013980", 759, "PSAE001c01", 3450, 11,
10, float(1.000000), (736.0 - 105.0 + 1) / 759, 105, 736)
m0 = ModelHit(h0, gene_gspd, GeneStatus.ACCESSORY)
m1 = ModelHit(h1, gene_gspd, GeneStatus.ACCESSORY)
l0 = Loner(h0, gene_ref=gene_gspd, gene_status=GeneStatus.ACCESSORY, counterpart=[m1])
l1 = Loner(h1, gene_ref=gene_gspd, gene_status=GeneStatus.ACCESSORY, counterpart=[m0])
l = get_best_hit_4_func(gene_name, [l0, l1], key='i_eval')
self.assertEqual(l, l0)
#################################
# based on the profile_coverage #
#################################
h0 = CoreHit(gene_gspd, "PSAE001c01_006940", 803, "PSAE001c01", 3450, 10,
10, 10, (741.0 - 104.0 + 1) / 803, 104, 741)
h1 = CoreHit(gene_gspd, "PSAE001c01_013980", 759, "PSAE001c01", 3450, 10,
10, 11, (736.0 - 105.0 + 1) / 759, 105, 736)
m0 = ModelHit(h0, gene_gspd, GeneStatus.ACCESSORY)
m1 = ModelHit(h1, gene_gspd, GeneStatus.ACCESSORY)
l0 = Loner(h0, gene_ref=gene_gspd, gene_status=GeneStatus.ACCESSORY, counterpart=[m1])
l1 = Loner(h1, gene_ref=gene_gspd, gene_status=GeneStatus.ACCESSORY, counterpart=[m0])
l = get_best_hit_4_func(gene_name, [l0, l1], key='profile_coverage')
self.assertEqual(l, l1)
# bad criterion
with self.assertRaises(MacsypyError) as ctx:
get_best_hits([l0, l1], key='nimportnaoik')
self.assertEqual('The criterion for Hits comparison nimportnaoik does not exist or is not available.\n'
'It must be either "score", "i_eval" or "profile_coverage".', str(ctx.exception))
def test_save_extract(self):
gene_name = "gspD"
gene = CoreGene(self.model_location, gene_name, self.profile_factory)
shutil.copy(
self.find_data("hmm", gene_name + self.cfg.res_search_suffix()),
self.cfg.working_dir())
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
report = GembaseHMMReport(gene, report_path, self.cfg)
report.extract()
report.save_extract()
extract_filename = gene_name + self.cfg.res_extract_suffix()
extract_path = os.path.join(self.cfg.working_dir(),
self.cfg.hmmer_dir(), extract_filename)
self.assertTrue(os.path.exists(extract_path))
self.assertTrue(os.path.isfile(extract_path))
hits = [
CoreHit(gene, "NC_xxxxx_xx_056141", 803, "NC_xxxxx_xx", 141,
float(2e-236), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741),
CoreHit(gene, "PSAE001c01_006940", 803, "PSAE001c01", 68,
float(1.2e-234), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741),
CoreHit(gene, "PSAE001c01_013980", 759, "PSAE001c01", 69,
float(3.7e-76), float(255.8), float(1.000000),
(736.0 - 105.0 + 1) / 759, 105, 736),
CoreHit(gene, "PSAE001c01_017350", 600, "PSAE001c01", 70,
float(3.2e-27), float(94.2), float(0.500000),
(506.0 - 226.0 + 1) / 600, 226, 506),
CoreHit(gene, "PSAE001c01_018920", 776, "PSAE001c01", 71,
float(6.1e-183), float(608.4), float(1.000000),
(606.0 - 48.0 + 1) / 776, 48, 606),
CoreHit(gene, "PSAE001c01_031420", 658, "PSAE001c01", 73,
float(1.8e-210), float(699.3), float(1.000000),
(614.0 - 55.0 + 1) / 658, 55, 614)
]
expected_extract_path = os.path.join(self.cfg.working_dir(),
'expected_extract')
with open(expected_extract_path, 'w') as expected_extract:
extract = """# gene: {name} extract from {path} hmm output
# profile length= {len_profile:d}
# i_evalue threshold= {i_evalue:.3f}
# coverage threshold= {cov:.3f}
# hit_id replicon_name position_hit hit_sequence_length gene_name gene_system i_eval score profile_coverage sequence_coverage begin end
""".format(name=gene.name,
path=report_path,
len_profile=len(gene.profile),
i_evalue=self.cfg.i_evalue_sel(),
cov=self.cfg.coverage_profile())
expected_extract.write(extract)
for h in hits:
expected_extract.write(str(h))
self.assertFileEqual(extract_path, expected_extract_path)
def test_best_hit(self):
gene_name = 'gspD'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
shutil.copy(
self.find_data("hmm", gene_name + self.cfg.res_search_suffix()),
self.cfg.working_dir())
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
report = GembaseHMMReport(c_gene, report_path, self.cfg)
self.assertIsNone(report.best_hit())
report.extract()
best_hit = report.best_hit()
hit_expected = CoreHit(c_gene, "NC_xxxxx_xx_056141", 803,
"NC_xxxxx_xx", 141, float(2e-236), float(779.2),
float(1.000000), (741.0 - 104.0 + 1) / 803, 104,
741)
self.assertEqual(hit_expected, best_hit)
def test_str(self):
gene_name = "gspD"
gene = CoreGene(self.model_location, gene_name, self.profile_factory)
hit_prop = {'id': "PSAE001c01_006940",
'hit_seq_len': 803,
'replicon_name': "PSAE001c01",
'position': 694,
'i_eval': float(1.2e-234),
'score': float(779.2),
'gene_name': gene.name,
'profil_coverage': float(1.0),
'sequence_coverage': float(638.000000),
'begin': 104,
'end': 741
}
hit = CoreHit(gene, hit_prop['id'], hit_prop['hit_seq_len'], hit_prop['replicon_name'],
hit_prop['position'], hit_prop['i_eval'], hit_prop['score'],
hit_prop['profil_coverage'], hit_prop['sequence_coverage'], hit_prop['begin'], hit_prop['end'])
s = "{id}\t{replicon_name}\t{position:d}\t{hit_seq_len:d}\t{gene_name}\t{i_eval:.3e}" \
"\t{score:.3f}\t{profil_coverage:.3f}\t{sequence_coverage:.3f}\t{begin:d}\t{end:d}\n".format(**hit_prop)
self.assertEqual(s, str(hit))
def test_str(self):
gene_name = 'gspD'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
shutil.copy(
self.find_data("hmm", gene_name + self.cfg.res_search_suffix()),
self.cfg.working_dir())
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
report = GembaseHMMReport(c_gene, report_path, self.cfg)
report.extract()
hits = [
CoreHit(c_gene, "NC_xxxxx_xx_056141", 803, "NC_xxxxx_xx", 141,
float(2e-236), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741),
CoreHit(c_gene, "PSAE001c01_006940", 803, "PSAE001c01", 68,
float(1.2e-234), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741),
CoreHit(c_gene, "PSAE001c01_013980", 759, "PSAE001c01", 69,
float(3.7e-76), float(255.8), float(1.000000),
(736.0 - 105.0 + 1) / 759, 105, 736),
CoreHit(c_gene, "PSAE001c01_017350", 600, "PSAE001c01", 70,
float(3.2e-27), float(94.2), float(0.500000),
(506.0 - 226.0 + 1) / 600, 226, 506),
CoreHit(c_gene, "PSAE001c01_018920", 776, "PSAE001c01", 71,
float(6.1e-183), float(608.4), float(1.000000),
(606.0 - 48.0 + 1) / 776, 48, 606),
CoreHit(c_gene, "PSAE001c01_031420", 658, "PSAE001c01", 73,
float(1.8e-210), float(699.3), float(1.000000),
(614.0 - 55.0 + 1) / 658, 55, 614)
]
s = f"# gene: {c_gene.name} extract from {report_path} hmm output\n"
s += f"# profile length= {len(c_gene.profile):d}\n"
s += f"# i_evalue threshold= {self.cfg.i_evalue_sel():.3f}\n"
s += f"# coverage threshold= {self.cfg.coverage_profile():.3f}\n"
s += "# hit_id replicon_name position_hit hit_sequence_length gene_name gene_system i_eval score " \
"profile_coverage sequence_coverage begin end\n"
for h in hits:
s += str(h)
self.assertMultiLineEqual(str(report), s)
def test_extract(self):
gene_name = "gspD"
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
shutil.copy(
self.find_data("hmm", gene_name + self.cfg.res_search_suffix()),
self.cfg.working_dir())
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
report = GeneralHMMReport(c_gene, report_path, self.cfg)
report.extract()
self.assertEqual(len(report.hits), 6)
# gene, model, hit_id, hit_seq_ length replicon_name, pos_hit, i_eval,
# score, profile_coverage, sequence_coverage, begin_match, end_match
hits = [
CoreHit(c_gene, "NC_xxxxx_xx_056141", 803, "Unordered", 141,
float(2e-236), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741),
CoreHit(c_gene, "PSAE001c01_006940", 803, "Unordered", 68,
float(1.2e-234), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741),
CoreHit(c_gene, "PSAE001c01_013980", 759, "Unordered", 69,
float(3.7e-76), float(255.8), float(1.000000),
(736.0 - 105.0 + 1) / 759, 105, 736),
CoreHit(c_gene, "PSAE001c01_017350", 600, "Unordered", 70,
float(3.2e-27), float(94.2), float(0.500000),
(506.0 - 226.0 + 1) / 600, 226, 506),
CoreHit(c_gene, "PSAE001c01_018920", 776, "Unordered", 71,
float(6.1e-183), float(608.4), float(1.000000),
(606.0 - 48.0 + 1) / 776, 48, 606),
CoreHit(c_gene, "PSAE001c01_031420", 658, "Unordered", 73,
float(1.8e-210), float(699.3), float(1.000000),
(614.0 - 55.0 + 1) / 658, 55, 614)
]
self.assertListEqual(hits, report.hits)
report = GeneralHMMReport(c_gene, report_path, self.cfg)
report.hits = hits
self.assertIsNone(report.extract())
def test_sort_hits_by_status(self):
ordered_match_maker = OrderedMatchMaker(self.model,
self.cfg.redundancy_penalty())
mandatory_exp = [self.m_hits['mh_sctn'], self.m_hits['mh_sctj']]
accessory_exp = [self.m_hits['mh_gspd']]
neutral_exp = [self.m_hits['mh_toto']]
forbidden_exp = [self.m_hits['mh_abc']]
mandatory, accessory, neutral, forbidden = ordered_match_maker.sort_hits_by_status(
mandatory_exp + accessory_exp + neutral_exp + forbidden_exp)
self.assertListEqual([h.gene.name for h in mandatory_exp],
[h.gene.name for h in mandatory])
self.assertListEqual([h.gene.name for h in accessory_exp],
[h.gene.name for h in accessory])
self.assertListEqual([h.gene.name for h in neutral_exp],
[h.gene.name for h in neutral])
self.assertListEqual([h.gene.name for h in forbidden_exp],
[h.gene.name for h in forbidden])
# do the same but with exchangeable
mandatory_exp_exch = [
self.m_hits['mh_sctn_flg'], self.m_hits['mh_sctj_flg']
]
accessory_exp_exch = [self.m_hits['mh_gspd_ex']]
neutral_exp_exch = [self.m_hits['mh_toto_ex']]
forbidden_exp_exch = [self.m_hits['mh_abc_ex']]
mandatory, accessory, neutral, forbidden = ordered_match_maker.sort_hits_by_status(
mandatory_exp_exch + accessory_exp_exch + neutral_exp_exch +
forbidden_exp_exch)
self.assertListEqual([h.gene.name for h in mandatory_exp_exch],
[h.gene.name for h in mandatory])
self.assertListEqual([h.gene.name for h in accessory_exp_exch],
[h.gene.name for h in accessory])
self.assertListEqual([h.gene.name for h in neutral_exp_exch],
[h.gene.name for h in neutral])
self.assertListEqual([h.gene.name for h in forbidden_exp_exch],
[h.gene.name for h in forbidden])
# test if gene_ref is the ModelGene
# alternate_of return the ModelGene of the function
self.assertListEqual(
[h.gene.name for h in mandatory_exp],
[h.gene_ref.alternate_of().name for h in mandatory])
self.assertListEqual(
[h.gene.name for h in accessory_exp],
[h.gene_ref.alternate_of().name for h in accessory])
self.assertListEqual([h.gene.name for h in neutral_exp],
[h.gene_ref.alternate_of().name for h in neutral])
self.assertListEqual(
[h.gene.name for h in forbidden_exp],
[h.gene_ref.alternate_of().name for h in forbidden])
# test if the hit does not refer to gene belonging to the model
model2 = Model("foo/model_B", 10)
cg_fliE = CoreGene(self.model_location, "fliE", self.profile_factory)
ch_fliE = CoreHit(cg_fliE, "hit_fliE", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
mg_fliE = ModelGene(cg_fliE, model2)
mh_fliE = ModelHit(ch_fliE, mg_fliE, GeneStatus.NEUTRAL)
with self.assertRaises(MacsypyError) as ctx:
with self.catch_log():
ordered_match_maker.sort_hits_by_status([mh_fliE])
self.assertEqual(str(ctx.exception),
"Gene 'fliE' not found in model 'foo/model_B'")
def test_LikelySystemSerializer_txt(self):
model = Model("foo/FOO", 10)
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
c_gene_sctn = CoreGene(self.model_location, "sctN",
self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model)
model.add_accessory_gene(gene_sctn)
c_gene_abc = CoreGene(self.model_location, "abc", self.profile_factory)
gene_abc = ModelGene(c_gene_abc, model)
model.add_forbidden_gene(gene_abc)
hit_1 = CoreHit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_1 = ModelHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = CoreHit(c_gene_sctj, "hit_2", 803, "replicon_id", 2, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_2 = ModelHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = CoreHit(c_gene_sctn, "hit_3", 803, "replicon_id", 3, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_3 = ModelHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
hit_4 = CoreHit(c_gene_abc, "hit_4", 803, "replicon_id", 4, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_hit_4 = ModelHit(hit_4, gene_abc, GeneStatus.FORBIDDEN)
ls_1 = LikelySystem(model, [v_hit_1], [v_hit_2, v_hit_3], [],
[v_hit_4])
hit_multi_sys_tracker = HitSystemTracker([ls_1])
ser = TxtLikelySystemSerializer()
txt = ser.serialize(ls_1, hit_multi_sys_tracker)
expected_txt = """This replicon contains genetic materials needed for system foo/FOO
WARNING there quorum is reached but there is also some forbidden genes.
system id = replicon_id_FOO_1
model = foo/FOO
replicon = replicon_id
hits = [('hit_1', 'gspD', 1), ('hit_2', 'sctJ', 2), ('hit_3', 'sctN', 3), ('hit_4', 'abc', 4)]
wholeness = 1.000
mandatory genes:
\t- gspD: 1 (gspD)
accessory genes:
\t- sctJ: 1 (sctJ)
\t- sctN: 1 (sctN)
neutral genes:
forbidden genes:
\t- abc: 1 (abc)
Use ordered replicon to have better prediction.
"""
self.assertEqual(txt, expected_txt)
def test_SolutionSerializer_tsv(self):
model_name = 'foo'
model_location = ModelLocation(
path=os.path.join(self.cfg.models_dir()[0], model_name))
###########
# Model B #
###########
model_B = Model("foo/B", 10)
c_gene_sctn_flg = CoreGene(model_location, "sctN_FLG",
self.profile_factory)
gene_sctn_flg = ModelGene(c_gene_sctn_flg, model_B)
c_gene_sctj_flg = CoreGene(model_location, "sctJ_FLG",
self.profile_factory)
gene_sctj_flg = ModelGene(c_gene_sctj_flg, model_B)
c_gene_flgB = CoreGene(model_location, "flgB", self.profile_factory)
gene_flgB = ModelGene(c_gene_flgB, model_B)
c_gene_tadZ = CoreGene(model_location, "tadZ", self.profile_factory)
gene_tadZ = ModelGene(c_gene_tadZ, model_B)
model_B.add_mandatory_gene(gene_sctn_flg)
model_B.add_mandatory_gene(gene_sctj_flg)
model_B.add_accessory_gene(gene_flgB)
model_B.add_accessory_gene(gene_tadZ)
###########
# Model A #
###########
model_A = Model("foo/A", 10)
c_gene_sctn = CoreGene(model_location, "sctN", self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model_A)
gene_sctn_hom = Exchangeable(c_gene_sctn_flg, gene_sctn)
gene_sctn.add_exchangeable(gene_sctn_hom)
c_gene_sctj = CoreGene(model_location, "sctJ", self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model_A)
gene_sctj_an = Exchangeable(c_gene_sctj_flg, gene_sctj)
gene_sctj.add_exchangeable(gene_sctj_an)
c_gene_gspd = CoreGene(model_location, "gspD", self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model_A)
gene_gspd_an = Exchangeable(c_gene_flgB, gene_gspd)
gene_gspd.add_exchangeable(gene_gspd_an)
c_gene_abc = CoreGene(model_location, "abc", self.profile_factory)
gene_abc = ModelGene(c_gene_abc, model_A, loner=True)
gene_abc_ho = Exchangeable(c_gene_tadZ, gene_abc)
gene_abc.add_exchangeable(gene_abc_ho)
model_A.add_mandatory_gene(gene_sctn)
model_A.add_mandatory_gene(gene_sctj)
model_A.add_accessory_gene(gene_gspd)
model_A.add_accessory_gene(gene_abc)
# CoreHit(gene, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
# pos score
h_sctj = CoreHit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_sctj = ModelHit(h_sctj, gene_sctj, GeneStatus.MANDATORY)
h_sctn = CoreHit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 2, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_sctn = ModelHit(h_sctn, gene_sctn, GeneStatus.MANDATORY)
h_gspd = CoreHit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 3, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_gspd = ModelHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
h_sctj_flg = CoreHit(c_gene_sctj_flg, "hit_sctj_flg", 803,
"replicon_id", 10, 1.0, 1.0, 1.0, 1.0, 10, 20)
h_flgB = CoreHit(c_gene_flgB, "hit_flgB", 803, "replicon_id", 11, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_abc = CoreHit(c_gene_abc, "hit_abc", 803, "replicon_id", 20, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_abc2 = CoreHit(c_gene_abc, "hit_abc2", 803, "replicon_id", 50, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_tadZ = CoreHit(c_gene_tadZ, "hit_tadZ", 803, "replicon_id", 40, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_sctj_flg = ModelHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY)
mh_flgB = ModelHit(h_flgB, gene_flgB, GeneStatus.ACCESSORY)
mh_abc = ModelHit(h_abc, gene_abc, GeneStatus.ACCESSORY)
mh_abc2 = ModelHit(h_abc2, gene_abc, GeneStatus.ACCESSORY)
mh_tadZ = ModelHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY)
model_A._min_mandatory_genes_required = 2
model_A._min_genes_required = 2
c1 = Cluster([mh_sctj, mh_sctn, mh_gspd], model_A, self.hit_weights)
c2 = Cluster([mh_sctj, mh_sctn], model_A, self.hit_weights)
c3 = Cluster([
Loner(h_abc,
gene_ref=gene_abc,
gene_status=GeneStatus.ACCESSORY,
counterpart=[mh_abc2])
], model_A, self.hit_weights)
model_B._min_mandatory_genes_required = 1
model_B._min_genes_required = 2
c5 = Cluster([mh_sctj_flg, mh_tadZ, mh_flgB], model_B,
self.hit_weights)
sys_A = System(model_A, [c1, c2, c3], self.cfg.redundancy_penalty())
# score = 2.5, 2 , 0.35 = 4.85 - (2 * 1.5) = 1.85
sys_A.id = "sys_id_A"
sys_B = System(model_B, [c5], self.cfg.redundancy_penalty())
# score = 2.0
sys_B.id = "sys_id_B"
sol = Solution([sys_A, sys_B])
sol_id = '12'
hit_multi_sys_tracker = HitSystemTracker([sys_A, sys_B])
sol_serializer = TsvSolutionSerializer()
sol_tsv = '\t'.join([
sol_id, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1', '1.000', '1.850', '2', 'sctJ', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctn', 'sctN', '2', 'foo/A',
'sys_id_A', '2', '1', '1.000', '1.850', '2', 'sctN', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_gspd', 'gspD', '3', 'foo/A',
'sys_id_A', '2', '1', '1.000', '1.850', '2', 'gspD', 'accessory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '2', '1.000', '1.850', '2', 'sctJ', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctn', 'sctN', '2', 'foo/A',
'sys_id_A', '2', '2', '1.000', '1.850', '2', 'sctN', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_abc', 'abc', '20', 'foo/A', 'sys_id_A',
'2', '-1', '1.000', '1.850', '2', 'abc', 'accessory', '803', '1.0',
'1.000', '1.000', '1.000', '10', '20', 'hit_abc2', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctj_flg', 'sctJ_FLG', '10', 'foo/B',
'sys_id_B', '1', '1', '0.750', '2.000', '1', 'sctJ_FLG',
'mandatory', '803', '1.0', '1.000', '1.000', '1.000', '10', '20',
'', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_flgB', 'flgB', '11', 'foo/B',
'sys_id_B', '1', '1', '0.750', '2.000', '1', 'flgB', 'accessory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_tadZ', 'tadZ', '40', 'foo/B',
'sys_id_B', '1', '1', '0.750', '2.000', '1', 'tadZ', 'accessory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
ser = sol_serializer.serialize(sol, sol_id, hit_multi_sys_tracker)
self.maxDiff = None
self.assertEqual(ser, sol_tsv)
def test_parse_hmm_body(self):
def make_hmm_group(hmm_string):
hmm_file = StringIO(hmm_string)
hmm_hits = (
x[1] for x in groupby(hmm_file, lambda l: l.startswith('>>')))
header = next(hmm_hits)
body = next(hmm_hits)
return body
gene_name = "gspD"
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
report_path = os.path.join(self.cfg.working_dir(),
gene_name + self.cfg.res_search_suffix())
report = GembaseHMMReport(c_gene, report_path, self.cfg)
# with one significant hit
hmm = """>> NC_xxxxx_xx_056141 C ATG TAA 6260390 6261757 Valid PA5567 1368 _NP_254254.1_ PA5567 1 6260390 6261757 | tRNA modific
# score bias c-Evalue i-Evalue hmmfrom hmm to alifrom ali to envfrom env to acc
--- ------ ----- --------- --------- ------- ------- ------- ------- ------- ------- ----
1 ! 779.2 5.5 1.4e-237 2e-236 1 596 [] 104 741 .. 104 741 .. 0.93
Alignments for each domain:
"""
body = make_hmm_group(hmm)
hits = report._parse_hmm_body('NC_xxxxx_xx_056141', 596, 803, 0.5,
'NC_xxxxx_xx', 141, 0.5, body)
expected_hits = [
CoreHit(c_gene, "NC_xxxxx_xx_056141", 803, "NC_xxxxx_xx", 141,
float(2e-236), float(779.2), float(1.000000),
(741.0 - 104.0 + 1) / 803, 104, 741)
]
self.assertListEqual(hits, expected_hits)
# with no significant hit
hmm = """>> PSAE001c01_051090 C ATG TGA 5675714 5677858 Valid pilQ 2145 _PA5040_NP_253727.1_ PA5040 1 5675714 5677858 | type 4 f
# score bias c-Evalue i-Evalue hmmfrom hmm to alifrom ali to envfrom env to acc
--- ------ ----- --------- --------- ------- ------- ------- ------- ------- ------- ----
1 ! 27.1 0.2 6.3e-10 6.6e-07 1 120 [. 286 402 .. 286 407 .. 0.86
2 ! 186.2 0.1 4.2e-58 4.3e-55 294 590 .. 405 709 .. 397 712 .. 0.84
Alignments for each domain:
"""
body = make_hmm_group(hmm)
hits = report._parse_hmm_body('NC_xxxxx_xx_056141', 596, 803, 0.5,
'NC_xxxxx_xx', 141, 0.5, body)
expected_hits = []
self.assertListEqual(hits, expected_hits)
# with no hit
hmm = """>> PSAE001c01_051090 C ATG TGA 5675714 5677858 Valid pilQ 2145 _PA5040_NP_253727.1_ PA5040 1 5675714 5677858 | type 4 f
bla bla
"""
body = make_hmm_group(hmm)
hits = report._parse_hmm_body('NC_xxxxx_xx_056141', 596, 803, 0.5,
'NC_xxxxx_xx', 141, 0.5, body)
expected_hits = []
self.assertListEqual(hits, expected_hits)
# with invalid hmm
hmm = """>> NC_xxxxx_xx_056141 C ATG TAA 6260390 6261757 Valid PA5567 1368 _NP_254254.1_ PA5567 1 6260390 6261757 | tRNA modific
# score bias c-Evalue i-Evalue hmmfrom hmm to alifrom ali to envfrom env to acc
--- ------ ----- --------- --------- ------- ------- ------- ------- ------- ------- ----
1 ! 779.2 5.5 1.4e-237 foo 1 596 [] 104 741 .. 104 741 .. 0.93
Alignments for each domain:
"""
body = make_hmm_group(hmm)
with self.assertRaises(ValueError) as ctx:
report._parse_hmm_body('NC_xxxxx_xx_056141', 596, 803, 0.5,
'NC_xxxxx_xx', 141, 0.5, body)
self.assertEqual(
str(ctx.exception),
"""Invalid line to parse : 1 ! 779.2 5.5 1.4e-237 foo 1 596 [] 104 741 .. 104 741 .. 0.93
:could not convert string to float: 'foo'""")
def test_SpecialHitSerializer_tsv(self):
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'gembase'
args.models_dir = self.find_data('models')
cfg = Config(MacsyDefaults(), args)
model_name = 'foo'
models_location = ModelLocation(
path=os.path.join(args.models_dir, model_name))
# we need to reset the ProfileFactory
# because it's a like a singleton
# so other tests are influenced by ProfileFactory and it's configuration
# for instance search_genes get profile without hmmer_exe
profile_factory = ProfileFactory(cfg)
model = Model("foo/T2SS", 10)
gene_name = "gspD"
cg_gspd = CoreGene(models_location, gene_name, profile_factory)
mg_gspd = ModelGene(cg_gspd, model, loner=True)
gene_name = "sctJ"
cg_sctj = CoreGene(models_location, gene_name, profile_factory)
mg_sctj = ModelGene(cg_sctj, model)
gene_name = "abc"
cg_abc = CoreGene(models_location, gene_name, profile_factory)
mg_abc = ModelGene(cg_abc, model)
model.add_mandatory_gene(mg_gspd)
model.add_accessory_gene(mg_sctj)
model.add_accessory_gene(mg_abc)
chit_abc = CoreHit(cg_abc, "hit_abc", 803, "replicon_id", 3, 1.0, 1.0,
1.0, 1.0, 10, 20)
chit_sctj = CoreHit(cg_sctj, "hit_sctj", 803, "replicon_id", 4, 1.0,
1.0, 1.0, 1.0, 10, 20)
chit_gspd1 = CoreHit(cg_gspd, "hit_gspd1", 803, "replicon_id", 20, 1.0,
2.0, 1.0, 1.0, 10, 20)
chit_gspd2 = CoreHit(cg_gspd, "hit_gspd2", 803, "replicon_id", 30, 1.0,
3.0, 1.0, 1.0, 10, 20)
mhit_abc = ModelHit(chit_abc, mg_abc, GeneStatus.ACCESSORY)
mhit_sctj = ModelHit(chit_sctj, mg_sctj, GeneStatus.ACCESSORY)
mhit_gspd1 = ModelHit(chit_gspd1, mg_gspd, GeneStatus.MANDATORY)
mhit_gspd2 = ModelHit(chit_gspd2, mg_gspd, GeneStatus.MANDATORY)
l_gspd1 = Loner(mhit_gspd1, counterpart=[mhit_gspd2])
l_gspd2 = Loner(mhit_gspd2, counterpart=[mhit_gspd1])
ser = TsvSpecialHitSerializer()
txt = ser.serialize([l_gspd1, l_gspd2])
expected_txt = "\t".join([
'replicon', 'model_fqn', 'function', 'gene_name', 'hit_id',
'hit_pos', 'hit_status', 'hit_seq_len', 'hit_i_eval', 'hit_score',
'hit_profile_cov', 'hit_seq_cov', 'hit_begin_match',
'hit_end_match'
])
expected_txt += "\n"
expected_txt += "\t".join([
'replicon_id', 'foo/T2SS', 'gspD', 'gspD', 'hit_gspd1', '20',
'mandatory', '803', '1.000e+00', '2.000', '1.000', '1.000', '10',
'20'
])
expected_txt += "\n"
expected_txt += "\t".join([
'replicon_id', 'foo/T2SS', 'gspD', 'gspD', 'hit_gspd2', '30',
'mandatory', '803', '1.000e+00', '3.000', '1.000', '1.000', '10',
'20'
])
expected_txt += "\n"
self.maxDiff = None
self.assertEqual(txt, expected_txt)
def test_SystemSerializer_str(self):
model_name = 'foo'
model_location = ModelLocation(
path=os.path.join(self.cfg.models_dir()[0], model_name))
model_A = Model("foo/A", 10)
model_B = Model("foo/B", 10)
c_gene_sctn_flg = CoreGene(model_location, "sctN_FLG",
self.profile_factory)
gene_sctn_flg = ModelGene(c_gene_sctn_flg, model_B)
c_gene_sctj_flg = CoreGene(model_location, "sctJ_FLG",
self.profile_factory)
gene_sctj_flg = ModelGene(c_gene_sctj_flg, model_B)
c_gene_flgB = CoreGene(model_location, "flgB", self.profile_factory)
c_gene_tadZ = CoreGene(model_location, "tadZ", self.profile_factory)
gene_tadZ = ModelGene(c_gene_tadZ, model_B)
c_gene_sctn = CoreGene(model_location, "sctN", self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model_A)
gene_sctn_hom = Exchangeable(c_gene_sctn_flg, gene_sctn)
gene_sctn.add_exchangeable(gene_sctn_hom)
c_gene_sctj = CoreGene(model_location, "sctJ", self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model_A)
gene_sctj_an = Exchangeable(c_gene_sctj_flg, gene_sctj)
gene_sctj.add_exchangeable(gene_sctj_an)
c_gene_gspd = CoreGene(model_location, "gspD", self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model_A)
gene_gspd_an = Exchangeable(c_gene_flgB, gene_gspd)
gene_gspd.add_exchangeable(gene_gspd_an)
c_gene_abc = CoreGene(model_location, "abc", self.profile_factory)
gene_abc = ModelGene(c_gene_abc, model_A)
gene_abc_ho = Exchangeable(c_gene_tadZ, gene_abc)
gene_abc.add_exchangeable(gene_abc_ho)
model_A.add_mandatory_gene(gene_sctn)
model_A.add_mandatory_gene(gene_sctj)
model_A.add_accessory_gene(gene_gspd)
model_A.add_forbidden_gene(gene_abc)
model_B.add_mandatory_gene(gene_sctn_flg)
model_B.add_mandatory_gene(gene_sctj_flg)
model_B.add_accessory_gene(gene_gspd)
model_B.add_accessory_gene(gene_tadZ)
h_sctj = CoreHit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_sctn = CoreHit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_gspd = CoreHit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_sctj_flg = CoreHit(c_gene_sctj_flg, "hit_sctj_flg", 803,
"replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
h_tadZ = CoreHit(c_gene_tadZ, "hit_tadZ", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20)
model_A._min_mandatory_genes_required = 2
model_A._min_genes_required = 2
c1 = Cluster([
ModelHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ModelHit(h_sctn, gene_sctn, GeneStatus.MANDATORY),
ModelHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_A, self.hit_weights)
c2 = Cluster([
ModelHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ModelHit(h_sctn, gene_sctn, GeneStatus.MANDATORY)
], model_A, self.hit_weights)
model_B._min_mandatory_genes_required = 1
model_B._min_genes_required = 2
c3 = Cluster([
ModelHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ModelHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ModelHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_B, self.hit_weights)
sys_A = System(model_A, [c1, c2], self.cfg.redundancy_penalty())
sys_A.id = "sys_id_A"
sys_B = System(model_B, [c3], self.cfg.redundancy_penalty())
sys_B.id = "sys_id_B"
hit_multi_sys_tracker = HitSystemTracker([sys_A, sys_B])
system_serializer = TxtSystemSerializer()
sys_str = f"""system id = {sys_A.id}
model = foo/A
replicon = replicon_id
clusters = [('hit_sctj', 'sctJ', 1), ('hit_sctn', 'sctN', 1), ('hit_gspd', 'gspD', 1)], [('hit_sctj', 'sctJ', 1), ('hit_sctn', 'sctN', 1)]
occ = 2
wholeness = 1.000
loci nb = 2
score = 1.500
mandatory genes:
\t- sctN: 2 (sctN, sctN)
\t- sctJ: 2 (sctJ, sctJ)
accessory genes:
\t- gspD: 1 (gspD [sys_id_B])
neutral genes:
"""
self.assertEqual(
sys_str, system_serializer.serialize(sys_A, hit_multi_sys_tracker))
def test_filter(self):
model_fqn = "foo/bar"
model = Model(model_fqn, 10)
model_2 = Model("foo/buz", 10)
gene_name = 'sctJ_FLG'
sctJ_FLG_core = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(sctJ_FLG_core, model)
model.add_mandatory_gene(sctJ_FLG)
gene_name = 'sctJ'
sctJ_core = CoreGene(self.model_location, gene_name, self.profile_factory)
sctj = Exchangeable(sctJ_core, sctJ_FLG)
sctJ_FLG.add_exchangeable(sctj)
gene_name = 'sctN_FLG'
sctN_FLG_core = CoreGene(self.model_location, gene_name, self.profile_factory)
sctN_FLG = ModelGene(sctN_FLG_core, model)
model.add_accessory_gene(sctN_FLG)
gene_name = 'sctN'
sctN_core = CoreGene(self.model_location, gene_name, self.profile_factory)
sctn = Exchangeable(sctN_core, sctN_FLG)
sctN_FLG.add_exchangeable(sctn)
gene_name = 'sctC'
sctC_core = CoreGene(self.model_location, gene_name, self.profile_factory)
sctC = ModelGene(sctC_core, model)
model.add_forbidden_gene(sctC)
gene_name = 'toto'
toto_core = CoreGene(self.model_location, gene_name, self.profile_factory)
toto = ModelGene(toto_core, model)
model.add_neutral_gene(toto)
gene_name = 'totote'
totote_core = CoreGene(self.model_location, gene_name, self.profile_factory)
totote = Exchangeable(totote_core, toto)
toto.add_exchangeable(totote)
gene_name = 'gspD'
gspd_core = CoreGene(self.model_location, gene_name, self.profile_factory)
gspd = ModelGene(gspd_core, model_2)
gene_name = 'tadZ'
tadz_core = CoreGene(self.model_location, gene_name, self.profile_factory)
tadz = Exchangeable(tadz_core, gspd)
gspd.add_exchangeable(tadz)
hit_to_keep = []
for gene in (sctJ_FLG, sctN_FLG, sctC, toto, totote):
hit_to_keep.append(CoreHit(gene,
f"PSAE001c01_{gene.name}",
1, "PSAE001c01", 1, 1.0, 1.0, 1.0, 1.0, 1, 2)
)
hit_to_filter_out = []
for gene in (gspd, tadz):
hit_to_filter_out.append(CoreHit(gene,
f"PSAE001c01_{gene.name}",
1, "PSAE001c01", 1, 1.0, 1.0, 1.0, 1.0, 1, 2)
)
filtered_hits = model.filter(hit_to_keep + hit_to_filter_out)
self.assertListEqual(sorted(hit_to_keep), sorted(filtered_hits))
def setUp(self) -> None:
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'gembase'
args.models_dir = self.find_data('models')
self.cfg = Config(MacsyDefaults(), args)
self.model_name = 'foo'
self.model_location = ModelLocation(
path=os.path.join(args.models_dir, self.model_name))
self.profile_factory = ProfileFactory(self.cfg)
self.model = Model("foo/model_A", 10)
c_gene_sctn = CoreGene(self.model_location, "sctN",
self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, self.model)
c_gene_sctn_flg = CoreGene(self.model_location, "sctN_FLG",
self.profile_factory)
gene_sctn_flg = Exchangeable(c_gene_sctn_flg, gene_sctn)
gene_sctn.add_exchangeable(gene_sctn_flg)
c_gene_sctj = CoreGene(self.model_location, "sctJ",
self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, self.model)
c_gene_sctj_flg = CoreGene(self.model_location, "sctJ_FLG",
self.profile_factory)
gene_sctj_flg = Exchangeable(c_gene_sctj_flg, gene_sctj)
gene_sctj.add_exchangeable(gene_sctj_flg)
c_gene_gspd = CoreGene(self.model_location, "gspD",
self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, self.model)
c_gene_flgb = CoreGene(self.model_location, "flgB",
self.profile_factory)
gene_gspd_ex = Exchangeable(c_gene_flgb, gene_gspd)
gene_gspd.add_exchangeable(gene_gspd_ex)
c_gene_abc = CoreGene(self.model_location, "abc", self.profile_factory)
gene_abc = ModelGene(c_gene_abc, self.model)
c_gene_tadz = CoreGene(self.model_location, "tadZ",
self.profile_factory)
gene_abc_ex = Exchangeable(c_gene_tadz, gene_abc)
gene_abc.add_exchangeable(gene_abc_ex)
c_gene_toto = CoreGene(self.model_location, "toto",
self.profile_factory)
gene_toto = ModelGene(c_gene_toto, self.model)
c_gene_totote = CoreGene(self.model_location, "totote",
self.profile_factory)
gene_toto_ex = Exchangeable(c_gene_totote, gene_toto)
gene_toto.add_exchangeable(gene_toto_ex)
self.model.add_mandatory_gene(gene_sctn)
self.model.add_mandatory_gene(gene_sctj)
self.model.add_accessory_gene(gene_gspd)
self.model.add_neutral_gene(gene_toto)
self.model.add_forbidden_gene(gene_abc)
self.c_hits = {
'ch_sctj':
CoreHit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20),
'ch_sctj_flg':
CoreHit(c_gene_sctj_flg, "hit_sctj_flg", 803, "replicon_id", 1,
1.0, 1.0, 1.0, 1.0, 10, 20),
'ch_sctn':
CoreHit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20),
'ch_sctn_flg':
CoreHit(c_gene_sctn_flg, "hit_sctn_flg", 803, "replicon_id", 1,
1.0, 1.0, 1.0, 1.0, 10, 20),
'ch_gspd':
CoreHit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20),
'ch_gspd_ex':
CoreHit(c_gene_flgb, "hit_gspd_an", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20),
'ch_abc':
CoreHit(c_gene_abc, "hit_abc", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20),
'ch_abc_ex':
CoreHit(c_gene_tadz, "hit_abc_ho", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20),
'ch_toto':
CoreHit(c_gene_toto, "hit_toto", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20),
'ch_toto_ex':
CoreHit(c_gene_totote, "hit_toto_ho", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20),
}
self.m_hits = {
'mh_sctj':
ModelHit(self.c_hits['ch_sctj'], gene_sctj, GeneStatus.MANDATORY),
'mh_sctj_flg':
ModelHit(self.c_hits['ch_sctj_flg'], gene_sctj_flg,
GeneStatus.MANDATORY),
'mh_sctn':
ModelHit(self.c_hits['ch_sctn'], gene_sctn, GeneStatus.MANDATORY),
'mh_sctn_flg':
ModelHit(self.c_hits['ch_sctn_flg'], gene_sctn_flg,
GeneStatus.MANDATORY),
'mh_gspd':
ModelHit(self.c_hits['ch_gspd'], gene_gspd, GeneStatus.ACCESSORY),
'mh_gspd_ex':
ModelHit(self.c_hits['ch_gspd_ex'], gene_gspd_ex,
GeneStatus.ACCESSORY),
'mh_abc':
ModelHit(self.c_hits['ch_abc'], gene_abc, GeneStatus.FORBIDDEN),
'mh_abc_ex':
ModelHit(self.c_hits['ch_abc_ex'], gene_abc_ex,
GeneStatus.FORBIDDEN),
'mh_toto':
ModelHit(self.c_hits['ch_toto'], gene_toto, GeneStatus.NEUTRAL),
'mh_toto_ex':
ModelHit(self.c_hits['ch_toto_ex'], gene_toto_ex,
GeneStatus.NEUTRAL)
}