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 = Hit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_1 = ValidHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = Hit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_2 = ValidHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = Hit(c_gene_sctn, "hit_3", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
rc = RejectedClusters(model, [
Cluster([v_hit_1, v_hit_2], model, self.hit_weights),
Cluster([v_hit_3], model, self.hit_weights)
], ["bla bla"])
self.assertEqual(rc.hits, [v_hit_1, v_hit_2, v_hit_3])
self.assertEqual(rc.reasons, ["bla bla"])
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 = Hit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_1 = ValidHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = Hit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_2 = ValidHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = Hit(c_gene_sctn, "hit_3", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(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 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 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.c_gene_gspd = CoreGene(models_location, gene_name, profile_factory)
self.gene_gspd = ModelGene(self.c_gene_gspd, model)
gene_name = "sctJ"
self.c_gene_sctj = CoreGene(models_location, gene_name, profile_factory)
self.gene_sctj = ModelGene(self.c_gene_sctj, model)
model.add_mandatory_gene(self.gene_gspd)
model.add_accessory_gene(self.gene_sctj)
self.hit_1 = Hit(self.c_gene_gspd, "hit_1", 803, "replicon_id", 2, 1.0, 1.0, 1.0, 1.0, 10, 20)
self.hit_2 = Hit(self.c_gene_sctj, "hit_2", 803, "replicon_id", 3, 1.0, 1.0, 1.0, 1.0, 10, 20)
def test_fulfilled_function(self):
model = Model("foo/T2SS", 11)
c_gene_1 = CoreGene(self.model_location, "gspD", self.profile_factory)
c_gene_2 = CoreGene(self.model_location, "sctC", self.profile_factory)
c_gene_3 = CoreGene(self.model_location, "sctJ", self.profile_factory)
c_gene_4 = CoreGene(self.model_location, "sctJ_FLG", self.profile_factory)
gene_1 = ModelGene(c_gene_1, model)
gene_2 = ModelGene(c_gene_2, model)
gene_3 = ModelGene(c_gene_3, model)
gene_4 = Exchangeable(c_gene_4, gene_3)
gene_3.add_exchangeable(gene_4)
# Hit(gene, model, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
h10 = Hit(c_gene_1, "h10", 10, "replicon_1", 10, 1.0, 10.0, 1.0, 1.0, 10, 20)
v_h10 = ValidHit(h10, gene_1, GeneStatus.MANDATORY)
h20 = Hit(c_gene_2, "h20", 10, "replicon_1", 20, 1.0, 20.0, 1.0, 1.0, 10, 20)
v_h20 = ValidHit(h20, gene_2, GeneStatus.MANDATORY)
c = Cluster([v_h10, v_h20], model, self.hit_weights)
self.assertTrue(c.fulfilled_function(gene_1))
self.assertFalse(c.fulfilled_function(gene_3))
h50 = Hit(c_gene_4, "h50", 10, "replicon_1", 50, 1.0, 50.0, 1.0, 1.0, 10, 20)
v_h50 = ValidHit(h50, gene_4, GeneStatus.ACCESSORY)
c = Cluster([v_h10, v_h50], model, self.hit_weights)
self.assertTrue(c.fulfilled_function(gene_3))
def test_contains(self):
model = Model("foo/T2SS", 11)
c_gene_1 = CoreGene(self.model_location, "gspD", self.profile_factory)
c_gene_2 = CoreGene(self.model_location, "sctC", self.profile_factory)
c_gene_3 = CoreGene(self.model_location, "sctJ", self.profile_factory)
gene_1 = ModelGene(c_gene_1, model)
gene_2 = ModelGene(c_gene_2, model)
gene_3 = ModelGene(c_gene_3, model)
# Hit(gene, model, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
h10 = Hit(c_gene_1, "h10", 10, "replicon_1", 10, 1.0, 10.0, 1.0, 1.0, 10, 20)
v_h10 = ValidHit(h10, gene_1, GeneStatus.MANDATORY)
h20 = Hit(c_gene_2, "h20", 10, "replicon_1", 20, 1.0, 20.0, 1.0, 1.0, 10, 20)
v_h20 = ValidHit(h20, gene_2, GeneStatus.MANDATORY)
h30 = Hit(c_gene_3, "h30", 10, "replicon_1", 30, 1.0, 30.0, 1.0, 1.0, 10, 20)
v_h30 = ValidHit(h30, gene_3, GeneStatus.ACCESSORY)
h50 = Hit(c_gene_3, "h50", 10, "replicon_1", 50, 1.0, 50.0, 1.0, 1.0, 10, 20)
v_h50 = ValidHit(h50, gene_3, GeneStatus.ACCESSORY)
c1 = Cluster([v_h10, v_h20, v_h50], model, self.hit_weights)
self.assertTrue(v_h10 in c1)
self.assertFalse(v_h30 in c1)
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 = Hit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_1 = ValidHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = Hit(c_gene_sctj, "hit_2", 803, "replicon_id", 2, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_2 = ValidHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = Hit(c_gene_sctn, "hit_3", 803, "replicon_id", 3, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(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_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 = Hit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_1 = ValidHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = Hit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_2 = ValidHit(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_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 = Hit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_1 = ValidHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = Hit(c_gene_sctj, "hit_2", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_2 = ValidHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = Hit(c_gene_sctn, "hit_3", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(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_str(self):
"""
"""
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(c_gene, model_foo)
gene_name = 'sctJ'
c_sctJ = CoreGene(self.model_location, gene_name, self.profile_factory)
homolog = Exchangeable(c_sctJ, sctJ_FLG)
sctJ_FLG.add_exchangeable(homolog)
gene_name = 'sctN'
c_sctN = CoreGene(self.model_location, gene_name, self.profile_factory)
analog = Exchangeable(c_sctN, sctJ_FLG)
sctJ_FLG.add_exchangeable(analog)
s = """name : sctJ_FLG
inter_gene_max_space: 10
exchangeables: sctJ, sctN"""
self.assertEqual(str(sctJ_FLG), s)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(c_gene,
model_foo,
loner=True,
multi_system=True,
inter_gene_max_space=10)
s = """name : sctJ_FLG
inter_gene_max_space: 10
loner
multi_system"""
self.assertEqual(str(sctJ_FLG), s)
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 = Hit(c_gene_gspd, "hit_1", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_1 = ValidHit(hit_1, gene_gspd, GeneStatus.MANDATORY)
hit_2 = Hit(c_gene_sctj, "hit_2", 803, "replicon_id", 2, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_2 = ValidHit(hit_2, gene_sctj, GeneStatus.ACCESSORY)
hit_3 = Hit(c_gene_sctn, "hit_3", 803, "replicon_id", 3, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(hit_3, gene_sctn, GeneStatus.ACCESSORY)
hit_4 = Hit(c_gene_abc, "hit_4", 803, "replicon_id", 4, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_4 = ValidHit(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 test_init(self):
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene_1 = ModelGene(c_gene, model_foo)
with self.assertRaises(MacsypyError) as ctx:
ModelGene(gene_1, model_foo)
self.assertEqual(str(ctx.exception),
"The ModeleGene gene argument must be a CoreGene not <class 'macsypy.gene.ModelGene'>.")
def test_SystemSerializer_tsv(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)
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)
model.add_accessory_gene(gene_sctn)
h_gspd = Hit(c_gene_gspd, "h_gspd", 803, "replicon_id", 10, 1.0, 1.0,
1.0, 1.0, 10, 20)
v_h_gspd = ValidHit(h_gspd, gene_gspd, GeneStatus.MANDATORY)
h_sctj = Hit(c_gene_sctj, "h_sctj", 803, "replicon_id", 20, 1.0, 1.0,
1.0, 1.0, 20, 30)
v_h_sctj = ValidHit(h_sctj, gene_sctj, GeneStatus.ACCESSORY)
h_sctn_flg = Hit(c_gene_sctn_flg, "h_sctn_flg", 803, "replicon_id", 30,
1.0, 1.0, 1.0, 1.0, 30, 40)
v_h_sctn_flg = ValidHit(h_sctn_flg, gene_sctn_flg,
GeneStatus.ACCESSORY)
c1 = Cluster([v_h_gspd, v_h_sctj], model, self.hit_weights)
c2 = Cluster([v_h_sctn_flg], model, self.hit_weights)
sys_multi_loci = System(model, [c1, c2], self.cfg.redundancy_penalty())
hit_multi_sys_tracker = HitSystemTracker([sys_multi_loci])
system_serializer = TsvSystemSerializer()
sys_tsv = "\t".join([
"replicon_id", "h_gspd", "gspD", "10", "foo/T2SS",
sys_multi_loci.id, "1", "1.000", "1.900", "1", "gspD", "mandatory",
"803", "1.0", "1.000", "1.000", "1.000", "10", "20", ""
])
sys_tsv += "\n"
sys_tsv += "\t".join([
"replicon_id", "h_sctj", "sctJ", "20", "foo/T2SS",
sys_multi_loci.id, "1", "1.000", "1.900", "1", "sctJ", "accessory",
"803", "1.0", "1.000", "1.000", "1.000", "20", "30", ""
])
sys_tsv += "\n"
sys_tsv += "\t".join([
"replicon_id", "h_sctn_flg", "sctN_FLG", "30", "foo/T2SS",
sys_multi_loci.id, "1", "1.000", "1.900", "1", "sctN", "accessory",
"803", "1.0", "1.000", "1.000", "1.000", "30", "40", ""
])
sys_tsv += "\n"
self.assertEqual(
sys_tsv,
system_serializer.serialize(sys_multi_loci, hit_multi_sys_tracker))
def test_hash(self):
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene_1 = ModelGene(c_gene, model_foo)
gene_2 = ModelGene(c_gene, model_foo)
self.assertTrue(isinstance(hash(gene_1), int))
self.assertEqual(hash(gene_1), hash(gene_1))
self.assertNotEqual(hash(gene_1), hash(gene_2))
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_multi_system(self):
model = Model("T2SS", 10)
gene_name = 'sctJ_FLG'
c_gene_ref = CoreGene(self.model_location, gene_name, self.profile_factory)
gene_ref = ModelGene(c_gene_ref, model)
gene_ref_multi_system = ModelGene(c_gene_ref, model, multi_system=True)
gene_name = 'sctJ'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
homolog_1 = Exchangeable(c_gene, gene_ref)
homolog_2 = Exchangeable(c_gene, gene_ref_multi_system)
self.assertFalse(homolog_1.multi_system)
self.assertTrue(homolog_2.multi_system)
def test_loner(self):
"""
test getter for loner property
"""
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(c_gene, model_foo)
self.assertFalse(sctJ_FLG.loner)
gene_name = 'sctJ'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ = ModelGene(c_gene, model_foo, loner=True)
self.assertTrue(sctJ.loner)
def test_merge(self):
model = Model("foo/T2SS", 11)
c_gene_1 = CoreGene(self.model_location, "gspD", self.profile_factory)
c_gene_2 = CoreGene(self.model_location, "sctC", self.profile_factory)
c_gene_3 = CoreGene(self.model_location, "sctJ", self.profile_factory)
gene_1 = ModelGene(c_gene_1, model)
gene_2 = ModelGene(c_gene_2, model)
gene_3 = ModelGene(c_gene_3, model)
# Hit(gene, model, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
h10 = Hit(c_gene_1, "h10", 10, "replicon_1", 10, 1.0, 10.0, 1.0, 1.0, 10, 20)
v_h10 = ValidHit(h10, gene_1, GeneStatus.MANDATORY)
h20 = Hit(c_gene_2, "h20", 10, "replicon_1", 20, 1.0, 20.0, 1.0, 1.0, 10, 20)
v_h20 = ValidHit(h20, gene_2, GeneStatus.MANDATORY)
h30 = Hit(c_gene_3, "h30", 10, "replicon_1", 30, 1.0, 30.0, 1.0, 1.0, 10, 20)
v_h30 = ValidHit(h30, gene_3, GeneStatus.ACCESSORY)
h50 = Hit(c_gene_3, "h50", 10, "replicon_1", 50, 1.0, 50.0, 1.0, 1.0, 10, 20)
v_h50 = ValidHit(h50, gene_3, GeneStatus.ACCESSORY)
c1 = Cluster([v_h10, v_h20], model, self.hit_weights)
c2 = Cluster([v_h30, v_h50], model, self.hit_weights)
c1.merge(c2)
self.assertListEqual(c1.hits, [v_h10, v_h20, v_h30, v_h50])
c1 = Cluster([v_h10, v_h20], model, self.hit_weights)
c2 = Cluster([v_h30, v_h50], model, self.hit_weights)
c2.merge(c1)
self.assertListEqual(c2.hits, [v_h30, v_h50, v_h10, v_h20])
c1 = Cluster([v_h10, v_h20], model, self.hit_weights)
c2 = Cluster([v_h30, v_h50], model, self.hit_weights)
c1.merge(c2, before=True)
self.assertListEqual(c1.hits, [v_h30, v_h50, v_h10, v_h20])
model_2 = Model("foo/T3SS", 11)
c_gene_3 = CoreGene(self.model_location, "sctJ", self.profile_factory)
gene_3 = ModelGene(c_gene_3, model)
h30 = Hit(c_gene_3, "h30", 10, "replicon_2", 30, 1.0, 30.0, 1.0, 1.0, 10, 20)
v_h30 = ValidHit(h30, gene_3, GeneStatus.ACCESSORY)
h50 = Hit(c_gene_3, "h50", 10, "replicon_2", 50, 1.0, 50.0, 1.0, 1.0, 10, 20)
v_h50 = ValidHit(h50, gene_3, GeneStatus.ACCESSORY)
c3 = Cluster([v_h30, v_h50], model_2, self.hit_weights)
with self.assertRaises(MacsypyError) as ctx:
c1.merge(c3)
self.assertEqual(str(ctx.exception), "Try to merge Clusters from different model")
def test_multi_system(self):
"""
test getter for multi_system property
"""
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(c_gene, model_foo)
self.assertFalse(sctJ_FLG.multi_system)
gene_name = 'sctJ'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ = ModelGene(c_gene, model_foo, multi_system=True)
self.assertTrue(sctJ.multi_system)
def test_execute_hmm_protected_path(self):
# create a hmmdir with space in name
self.cfg.hmmer_dir = lambda: 'hmmer results'
# create sequence_db path with space in path
seq_path = os.path.join(self.cfg.working_dir(), "test test1.fasta")
shutil.copyfile(self.find_data("base", "test_1.fasta"), seq_path)
self.cfg._set_sequence_db(seq_path)
model = Model("foo/T2SS", 10)
gene_name = 'T5aSS_PF03797'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene = ModelGene(c_gene, model)
# case GA threshold in profile
profile_path = self.model_location.get_profile("T5aSS_PF03797")
profile = Profile(gene, self.cfg, profile_path)
report = profile.execute()
hmmer_raw_out = profile.hmm_raw_output
with open(hmmer_raw_out, 'r') as hmmer_raw_out_file:
first_l = hmmer_raw_out_file.readline()
# a hmmsearch output file has been produced
self.assertTrue(
first_l.startswith(
"# hmmsearch :: search profile(s) against a sequence database"
))
for i in range(5):
# skip 4 lines
l = hmmer_raw_out_file.readline()
# a hmmsearch used the abc profile line should become with: "# query HMM file: {the path tp hmm profile used}"
self.assertTrue(l.find(profile_path) != -1)
for i in range(3):
# skip 2 lines
l = hmmer_raw_out_file.readline()
self.assertEqual("# model-specific thresholding: GA cutoffs",
l.strip())
def test_is_Forbidden(self):
"""
test if gene belong to model mandatory genes
"""
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(c_gene, model_foo)
model_foo.add_mandatory_gene(sctJ_FLG)
self.assertFalse(sctJ_FLG.is_forbidden(model_foo))
gene_name = 'sctJ'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ = ModelGene(c_gene, model_foo)
model_foo.add_forbidden_gene(sctJ)
self.assertTrue(sctJ.is_forbidden(model_foo))
def test_execute_hmm_w_GA_n_nocutga(self):
# case GA threshold in profile but --no-cut-ga is set
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'gembase'
args.models_dir = self.find_data('models')
args.res_search_dir = tempfile.gettempdir()
args.log_level = 0
args.e_value_search = 0.5
args.no_cut_ga = True
cfg = Config(MacsyDefaults(), args)
model = Model("foo/T2SS", 10)
gene_name = 'T5aSS_PF03797'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene = ModelGene(c_gene, model)
profile_path = self.model_location.get_profile("T5aSS_PF03797")
profile = Profile(gene, cfg, profile_path)
report = profile.execute()
hmmer_raw_out = profile.hmm_raw_output
with open(hmmer_raw_out, 'r') as hmmer_raw_out_file:
for i in range(9):
l = hmmer_raw_out_file.readline()
self.assertEqual(
"# sequence reporting threshold: E-value <= 0.5", l.strip())
def test_get_loners(self):
model = Model("foo/T2SS", 11)
# handle name, topology type, and min/max positions in the sequence dataset for a replicon and list of genes.
# each genes is representing by a tuple (seq_id, length)"""
rep_info = RepliconInfo('linear', 1, 60, [(f"g_{i}", i * 10) for i in range(1, 7)])
core_genes = []
model_genes = []
for g_name in ('gspD', 'sctC', 'sctJ', 'sctN', 'abc'):
core_gene = CoreGene(self.model_location, g_name, self.profile_factory)
core_genes.append(core_gene)
model_genes.append(ModelGene(core_gene, model))
model_genes[3]._loner = True
model_genes[4]._loner = True
model.add_mandatory_gene(model_genes[0])
model.add_mandatory_gene(model_genes[1])
model.add_accessory_gene(model_genes[2])
model.add_accessory_gene(model_genes[3])
model.add_neutral_gene(model_genes[4])
# Hit(gene, model, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
h10 = Hit(core_genes[0], "h10", 10, "replicon_1", 10, 1.0, 10.0, 1.0, 1.0, 10, 20)
h20 = Hit(core_genes[1], "h20", 10, "replicon_1", 20, 1.0, 20.0, 1.0, 1.0, 10, 20)
h30 = Hit(core_genes[2], "h30", 10, "replicon_1", 30, 1.0, 30.0, 1.0, 1.0, 10, 20)
h61 = Hit(core_genes[3], "h61", 10, "replicon_1", 60, 1.0, 61.0, 1.0, 1.0, 10, 20)
h80 = Hit(core_genes[4], "h80", 10, "replicon_1", 80, 1.0, 80.0, 1.0, 1.0, 10, 20)
# loners are clusters of one hit
loners = get_loners([h10, h20, h30, h61, h80], model, self.hit_weights)
hit_from_clusters = [h.hits[0] for h in loners]
self.assertListEqual(hit_from_clusters, [h61, h80])
def test_execute_hmmer_failed(self):
fake_hmmer = os.path.join(tempfile.gettempdir(), 'hmmer_failed')
with open(fake_hmmer, 'w') as hmmer:
hmmer.write("""#! {}
import sys
sys.exit(127)
""".format(sysconfig.sys.executable))
try:
os.chmod(hmmer.name, 0o755)
self.cfg._options['hmmer'] = hmmer.name
model = Model("foo/T2SS", 10)
gene_name = 'abc'
c_gene = CoreGene(self.model_location, gene_name,
self.profile_factory)
gene = ModelGene(c_gene, model)
path = self.model_location.get_profile("abc", )
profile = Profile(gene, self.cfg, path)
with self.catch_log():
with self.assertRaisesRegex(
RuntimeError, "an error occurred during Hmmer "
"execution: command = .* : return code = 127 .*"
) as ctx:
profile.execute()
finally:
try:
os.unlink(fake_hmmer)
except Exception:
pass
def test_execute_hmm_wo_GA(self):
# case cut-ga but no GA threshold in hmmprofile
model = Model("foo/T2SS", 10)
gene_name = 'abc'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene = ModelGene(c_gene, model)
# case -cut-ga and GA threshold in profile
profile_path = self.model_location.get_profile("abc")
profile = Profile(gene, self.cfg, profile_path)
with self.catch_log() as log:
report = profile.execute()
hmmer_raw_out = profile.hmm_raw_output
with open(hmmer_raw_out, 'r') as hmmer_raw_out_file:
first_l = hmmer_raw_out_file.readline()
# a hmmsearch output file has been produced
self.assertTrue(
first_l.startswith(
"# hmmsearch :: search profile(s) against a sequence database"
))
for i in range(5):
# skip 4 lines
l = hmmer_raw_out_file.readline()
# a hmmsearch used the abc profile line should become with: "# query HMM file: {the path tp hmm profile used}"
self.assertTrue(l.find(profile_path) != -1)
for i in range(3):
# skip 2 lines
l = hmmer_raw_out_file.readline()
self.assertEqual(
'# sequence reporting threshold: E-value <= 0.1', l.strip())
def test_execute_hmm_with_GA(self):
for db_type in ("gembase", "ordered_replicon", "unordered"):
self.cfg._set_db_type(db_type)
model = Model("foo/T2SS", 10)
gene_name = 'T5aSS_PF03797'
c_gene = CoreGene(self.model_location, gene_name,
self.profile_factory)
gene = ModelGene(c_gene, model)
# case GA threshold in profile
profile_path = self.model_location.get_profile("T5aSS_PF03797")
profile = Profile(gene, self.cfg, profile_path)
report = profile.execute()
hmmer_raw_out = profile.hmm_raw_output
with open(hmmer_raw_out, 'r') as hmmer_raw_out_file:
first_l = hmmer_raw_out_file.readline()
# a hmmsearch output file has been produced
self.assertTrue(
first_l.startswith(
"# hmmsearch :: search profile(s) against a sequence database"
))
for i in range(5):
# skip 4 lines
l = hmmer_raw_out_file.readline()
# a hmmsearch used the abc profile line should become with: "# query HMM file: {the path tp hmm profile used}"
self.assertTrue(l.find(profile_path) != -1)
for i in range(3):
# skip 2 lines
l = hmmer_raw_out_file.readline()
self.assertEqual(
"# model-specific thresholding: GA cutoffs", l.strip())
# test if profile is executed only once per run
report_bis = profile.execute()
self.assertIs(report, report_bis)
def test_ga_threshold(self):
model = Model("foo/T2SS", 10)
gene_name = 'abc'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene = ModelGene(c_gene, model)
path = self.model_location.get_profile("abc")
profile = Profile(gene, self.cfg, path)
self.assertFalse(profile.ga_threshold)
gene_name = 'T5aSS_PF03797'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene = ModelGene(c_gene, model)
path = self.model_location.get_profile("T5aSS_PF03797")
profile = Profile(gene, self.cfg, path)
self.assertTrue(profile.ga_threshold)
def test_inter_gene_max_space(self):
"""
test getter for inter_gene_max_space property
"""
system_inter_gene_max_space = 40
gene_inter_gene_max_space = 50
model_foo = Model("foo", system_inter_gene_max_space)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ_FLG = ModelGene(c_gene, model_foo)
self.assertIsNone(sctJ_FLG.inter_gene_max_space, None)
gene_name = 'sctJ'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
sctJ = ModelGene(c_gene, model_foo, inter_gene_max_space=gene_inter_gene_max_space)
self.assertEqual(sctJ.inter_gene_max_space, gene_inter_gene_max_space)
def test_unknown_attribute(self):
model_foo = Model("foo", 10)
gene_name = 'sctJ_FLG'
c_gene = CoreGene(self.model_location, gene_name, self.profile_factory)
gene = ModelGene(c_gene, model_foo)
with self.assertRaises(AttributeError) as ctx:
gene.foo
self.assertEqual(str(ctx.exception), "'ModelGene' object has no attribute 'foo'")
def test_len(self):
model = Model("foo/T2SS", 11)
c_gene_1 = CoreGene(self.model_location, "gspD", self.profile_factory)
c_gene_2 = CoreGene(self.model_location, "sctC", self.profile_factory)
gene_1 = ModelGene(c_gene_1, model)
gene_2 = ModelGene(c_gene_2, model)
# Hit(gene, model, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
h10 = Hit(c_gene_1, "h10", 10, "replicon_1", 10, 1.0, 10.0, 1.0, 1.0, 10, 20)
v_h10 = ValidHit(h10, gene_1, GeneStatus.MANDATORY)
h20 = Hit(c_gene_2, "h20", 10, "replicon_1", 20, 1.0, 20.0, 1.0, 1.0, 10, 20)
v_h20 = ValidHit(h20, gene_2, GeneStatus.MANDATORY)
c1 = Cluster([v_h10, v_h20], model, self.hit_weights)
self.assertEqual(len(c1), 2)