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_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 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_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_eq(self):
v_hit_0 = ValidHit(self.hit_1, self.gene_sctj, GeneStatus.MANDATORY)
v_hit_1 = ValidHit(self.hit_1, self.gene_sctj, GeneStatus.MANDATORY)
v_hit_2 = ValidHit(self.hit_2, self.gene_gspd, GeneStatus.ACCESSORY)
self.assertEqual(v_hit_0, v_hit_1)
self.assertNotEqual(v_hit_0, v_hit_2)
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_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_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_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_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_init(self):
v_hit_1 = ValidHit(self.hit_1, self.gene_gspd, GeneStatus.MANDATORY)
self.assertEqual(v_hit_1.gene_ref, self.gene_gspd)
self.assertEqual(v_hit_1.status, GeneStatus.MANDATORY)
v_hit_2 = ValidHit(self.hit_2, self.gene_gspd, GeneStatus.ACCESSORY)
self.assertEqual(v_hit_2.gene_ref, self.gene_gspd)
self.assertEqual(v_hit_2.status, GeneStatus.ACCESSORY)
with self.assertRaises(MacsypyError) as ctx:
ValidHit(self.hit_1, self.c_gene_gspd, GeneStatus.MANDATORY)
self.assertEqual(str(ctx.exception),
"The ValidHit 'gene_ref' argument must be a ModelGene not <class 'macsypy.gene.CoreGene'>.")
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)
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 = 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.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_str(self):
model = Model("foo/T2SS", 11)
c_gene_1 = CoreGene(self.model_location, "gspD", self.profile_factory)
gene_1 = ModelGene(c_gene_1, model)
model.add_mandatory_gene(gene_1)
c_gene_2 = CoreGene(self.model_location, "sctC", self.profile_factory)
gene_2 = ModelGene(c_gene_2, model)
model.add_accessory_gene(gene_2)
# 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.ACCESSORY)
h40 = Hit(c_gene_1, "h40", 10, "replicon_1", 40, 1.0, 10.0, 1.0, 1.0,
10, 20)
v_h40 = ValidHit(h40, gene_1, GeneStatus.MANDATORY)
h50 = Hit(c_gene_2, "h50", 10, "replicon_1", 50, 1.0, 20.0, 1.0, 1.0,
10, 20)
v_h50 = ValidHit(h50, gene_2, GeneStatus.ACCESSORY)
c1 = Cluster([v_h10, v_h20], model, self.hit_weights)
c2 = Cluster([v_h40, v_h50], model, self.hit_weights)
r_c = RejectedClusters(model, [c1, c2], ["bla"])
expected_str = """Cluster:
- model = T2SS
- replicon = replicon_1
- hits = (h10, gspD, 10), (h20, sctC, 20)
Cluster:
- model = T2SS
- replicon = replicon_1
- hits = (h40, gspD, 40), (h50, sctC, 50)
These clusters have been rejected because:
\t- bla
"""
self.assertEqual(expected_str, str(r_c))
def test_init(self):
model = Model("foo/T2SS", 11)
c_gene_1 = CoreGene(self.model_location, "gspD", self.profile_factory)
gene_1 = ModelGene(c_gene_1, model)
model.add_mandatory_gene(gene_1)
c_gene_2 = CoreGene(self.model_location, "sctC", self.profile_factory)
gene_2 = ModelGene(c_gene_2, model)
model.add_accessory_gene(gene_2)
# 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.ACCESSORY)
c1 = Cluster([v_h10, v_h20], model, self.hit_weights)
r_c = RejectedClusters(model, c1, ["bla"])
self.assertListEqual(r_c.clusters, [c1])
self.assertEqual(r_c.reasons, ['bla'])
def test_init(self):
model_1 = 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_1)
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_1, GeneStatus.MANDATORY)
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_1, 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_1, GeneStatus.ACCESSORY)
with self.assertRaises(MacsypyError) as ctx:
with self.catch_log():
Cluster([v_h10, v_h20, v_h30, v_h50], model_1, self.hit_weights)
msg = "Cannot build a cluster from hits coming from different replicons"
self.assertEqual(str(ctx.exception), msg)
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_systems_to_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"
c_gene_gspd = CoreGene(models_location, gene_name, profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
gene_name = "sctJ"
c_gene_sctj = CoreGene(models_location, gene_name, 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)
system_1 = System(model, [
Cluster([v_hit_1, v_hit_2], model, HitWeight(**cfg.hit_weights()))
], cfg.redundancy_penalty())
system_tsv = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Systems found:
"""
system_tsv += "\t".join([
"replicon", "hit_id", "gene_name", "hit_pos", "model_fqn",
"sys_id", "sys_loci", "sys_wholeness", "sys_score", "sys_occ",
"hit_gene_ref", "hit_status", "hit_seq_len", "hit_i_eval",
"hit_score", "hit_profile_cov", "hit_seq_cov", "hit_begin_match",
"hit_end_match", "used_in"
])
system_tsv += "\n"
system_tsv += "\t".join([
"replicon_id", "hit_1", "gspD", "1", "foo/T2SS", system_1.id, "1",
"1.000", "1.500", "1", "gspD", "mandatory", "803", "1.0", "1.000",
"1.000", "1.000", "10", "20", ""
])
system_tsv += "\n"
system_tsv += "\t".join([
"replicon_id", "hit_2", "sctJ", "1", "foo/T2SS", system_1.id, "1",
"1.000", "1.500", "1", "sctJ", "accessory", "803", "1.0", "1.000",
"1.000", "1.000", "10", "20", ""
])
system_tsv += "\n\n"
f_out = StringIO()
track_multi_systems_hit = HitSystemTracker([system_1])
systems_to_tsv([system_1], track_multi_systems_hit, f_out)
self.assertMultiLineEqual(system_tsv, f_out.getvalue())
# test No system found
system_str = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# No Systems found
"""
f_out = StringIO()
track_multi_systems_hit = HitSystemTracker([])
systems_to_tsv([], track_multi_systems_hit, f_out)
self.assertMultiLineEqual(system_str, f_out.getvalue())
def test_solutions_to_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_A = Model("foo/A", 10)
model_B = Model("foo/B", 10)
model_C = Model("foo/C", 10)
c_gene_sctn_flg = CoreGene(models_location, "sctN_FLG",
profile_factory)
gene_sctn_flg = ModelGene(c_gene_sctn_flg, model_B)
c_gene_sctj_flg = CoreGene(models_location, "sctJ_FLG",
profile_factory)
gene_sctj_flg = ModelGene(c_gene_sctj_flg, model_B)
c_gene_flgB = CoreGene(models_location, "flgB", profile_factory)
gene_flgB = ModelGene(c_gene_flgB, model_B)
c_gene_tadZ = CoreGene(models_location, "tadZ", profile_factory)
gene_tadZ = ModelGene(c_gene_tadZ, model_B)
c_gene_sctn = CoreGene(models_location, "sctN", 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(models_location, "sctJ", 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(models_location, "gspD", 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(models_location, "abc", 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_flgB)
model_B.add_accessory_gene(gene_tadZ)
model_C.add_mandatory_gene(gene_sctn_flg)
model_C.add_mandatory_gene(gene_sctj_flg)
model_C.add_mandatory_gene(gene_flgB)
model_C.add_accessory_gene(gene_tadZ)
model_C.add_accessory_gene(gene_gspd)
h_sctj = Hit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_sctn = Hit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_gspd = Hit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_sctj_flg = Hit(c_gene_sctj_flg, "hit_sctj_flg", 803, "replicon_id",
1, 1.0, 1.0, 1.0, 1.0, 10, 20)
h_flgB = Hit(c_gene_flgB, "hit_flgB", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_tadZ = Hit(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
hit_weights = HitWeight(**cfg.hit_weights())
c1 = Cluster([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.MANDATORY),
ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_A, hit_weights)
c2 = Cluster([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.MANDATORY)
], model_A, hit_weights)
model_B._min_mandatory_genes_required = 1
model_B._min_genes_required = 2
c3 = Cluster([
ValidHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ValidHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ValidHit(h_flgB, gene_flgB, GeneStatus.ACCESSORY)
], model_B, hit_weights)
model_C._min_mandatory_genes_required = 1
model_C._min_genes_required = 2
c4 = Cluster([
ValidHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ValidHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ValidHit(h_flgB, gene_flgB, GeneStatus.MANDATORY),
ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_C, hit_weights)
sys_A = System(model_A, [c1, c2], cfg.redundancy_penalty())
sys_A.id = "sys_id_A"
sys_B = System(model_B, [c3], cfg.redundancy_penalty())
sys_B.id = "sys_id_B"
sys_C = System(model_C, [c4], cfg.redundancy_penalty())
sys_C.id = "sys_id_C"
sol_1 = [sys_A, sys_B]
sol_2 = [sys_A, sys_C]
sol_id_1 = '1'
sol_id_2 = '2'
sol_tsv = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Systems found:
"""
sol_tsv += "\t".join([
"sol_id", "replicon", "hit_id", "gene_name", "hit_pos",
"model_fqn", "sys_id", "sys_loci", "sys_wholeness", "sys_score",
"sys_occ", "hit_gene_ref", "hit_status", "hit_seq_len",
"hit_i_eval", "hit_score", "hit_profile_cov", "hit_seq_cov",
"hit_begin_match", "hit_end_match", "used_in"
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctJ', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_sctn', 'sctN', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctN', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_gspd', 'gspD', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'gspD', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctJ', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_sctn', 'sctN', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctN', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_sctj_flg', 'sctJ_FLG', '1', 'foo/B',
'sys_id_B', '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_1, 'replicon_id', 'hit_tadZ', 'tadZ', '1', 'foo/B',
'sys_id_B', '1', '0.750', '2.000', '1', 'tadZ', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_1, 'replicon_id', 'hit_flgB', 'flgB', '1', 'foo/B',
'sys_id_B', '1', '0.750', '2.000', '1', 'flgB', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctJ', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_sctn', 'sctN', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctN', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_gspd', 'gspD', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'gspD', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctJ', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_sctn', 'sctN', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctN', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_sctj_flg', 'sctJ_FLG', '1', 'foo/C',
'sys_id_C', '1', '0.800', '3.000', '1', 'sctJ_FLG', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', 'sys_id_B'
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_tadZ', 'tadZ', '1', 'foo/C',
'sys_id_C', '1', '0.800', '3.000', '1', 'tadZ', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', 'sys_id_B'
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_flgB', 'flgB', '1', 'foo/C',
'sys_id_C', '1', '0.800', '3.000', '1', 'flgB', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', 'sys_id_B'
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id_2, 'replicon_id', 'hit_gspd', 'gspD', '1', 'foo/C',
'sys_id_C', '1', '0.800', '3.000', '1', 'gspD', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', 'sys_id_A'
])
sol_tsv += "\n"
sol_tsv += "\n"
f_out = StringIO()
hit_multi_sys_tracker = HitSystemTracker([sys_A, sys_B])
solutions_to_tsv([sol_1, sol_2], hit_multi_sys_tracker, f_out)
self.assertMultiLineEqual(sol_tsv, f_out.getvalue())
def test_delegation(self):
v_hit_1 = ValidHit(self.hit_1, self.gene_gspd, GeneStatus.MANDATORY)
self.assertEqual(v_hit_1.get_position(), 2)
def test_score(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_tadZ = CoreGene(self.model_location, "tadZ", self.profile_factory)
gene_tadZ = ModelGene(c_gene_tadZ, model)
model.add_mandatory_gene(gene_tadZ)
c_gene_sctj = CoreGene(self.model_location, "sctC", self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
c_gene_sctJ_FLG = CoreGene(self.model_location, "sctJ_FLG", self.profile_factory)
analog_sctJ_FLG = Exchangeable(c_gene_sctJ_FLG, gene_sctj)
gene_sctj.add_exchangeable(analog_sctJ_FLG)
model.add_accessory_gene(gene_sctj)
c_gene_sctn = CoreGene(self.model_location, "sctN", self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model, loner=True)
c_gene_sctn_FLG = CoreGene(self.model_location, "sctN_FLG", self.profile_factory)
homolog_sctn_FLG = Exchangeable(c_gene_sctn_FLG, gene_sctn)
gene_sctn.add_exchangeable(homolog_sctn_FLG)
model.add_accessory_gene(gene_sctn)
c_gene_toto = CoreGene(self.model_location, "toto", self.profile_factory)
gene_toto = ModelGene(c_gene_toto, model)
model.add_neutral_gene(gene_toto)
c_gene_flie = CoreGene(self.model_location, "fliE", self.profile_factory)
gene_flie = ModelGene(c_gene_flie, model, loner=True, multi_system=True)
model.add_mandatory_gene(gene_flie)
h_gspd = Hit(c_gene_gspd, "h_gspd", 10, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_gspd = ValidHit(h_gspd, gene_gspd, GeneStatus.MANDATORY)
h_tadz = Hit(c_gene_tadZ, "h_tadz", 20, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_tadz = ValidHit(h_tadz, gene_tadZ, GeneStatus.MANDATORY)
h_sctj = Hit(c_gene_sctj, "h_sctj", 30, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_sctj = ValidHit(h_sctj, gene_sctj, GeneStatus.ACCESSORY)
h_sctj_an = Hit(c_gene_sctJ_FLG, "h_sctj_an", 30, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_sctj_an = ValidHit(h_sctj_an, analog_sctJ_FLG, GeneStatus.ACCESSORY)
h_sctn = Hit(c_gene_sctn, "sctn", 40, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_sctn = ValidHit(h_sctn, gene_sctn, GeneStatus.ACCESSORY)
h_sctn_hom = Hit(c_gene_sctn_FLG, "h_scth_hom", 30, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_sctn_hom = ValidHit(h_sctn_hom, homolog_sctn_FLG, GeneStatus.ACCESSORY)
h_toto = Hit(c_gene_sctn, "toto", 50, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_toto = ValidHit(h_toto, gene_toto, GeneStatus.NEUTRAL)
h_flie = Hit(c_gene_flie, "h_flie", 100, "replicon_id", 1, 1.0, 1.0, 1.0, 1.0, 10, 20)
v_h_flie = ValidHit(h_flie, gene_flie, GeneStatus.MANDATORY)
# 2 mandatory, 2 accessory no analog/homolog
c1 = Cluster([v_h_gspd, v_h_tadz, v_h_sctj, v_h_sctn], model, self.hit_weights)
self.assertEqual(c1.score, 3.0)
# 2 mandatory, 2 accessory 1 neutral, no analog/homolog
c1 = Cluster([v_h_gspd, v_h_tadz, v_h_sctj, v_h_sctn, v_h_toto], model, self.hit_weights)
self.assertEqual(c1.score, 3.0)
# 1 mandatory + 1 mandatory duplicated 1 time
# 1 accessory + 1 accessory duplicated 1 times
# no analog/homolog
c1 = Cluster([v_h_gspd, v_h_tadz, v_h_sctj, v_h_sctn, v_h_gspd, v_h_sctn], model, self.hit_weights)
self.assertEqual(c1.score, 3.0)
# 2 mandatory
# 1 accessory + 1 accessory homolog
c1 = Cluster([v_h_gspd, v_h_tadz, v_h_sctj, v_h_sctn_hom], model, self.hit_weights)
self.assertEqual(c1.score, 2.9)
# # 2 mandatory
# # 1 accessory + 1 accessory analog of the 1rst accessory
# c1 = Cluster([v_h_gspd, v_h_tadz, v_h_sctj, v_h_sctj_an], model, self.hit_weights)
# self.assertEqual(c1.score, 2.5)
# test loners multi system
c1 = Cluster([v_h_flie], model, self.hit_weights)
self.assertEqual(c1.score, 0.7)
# test the cache score
self.assertEqual(c1.score, 0.7)
non_valid_hit = ValidHit(h_sctn, gene_sctn, GeneStatus.FORBIDDEN)
c1 = Cluster([v_h_gspd, non_valid_hit, v_h_tadz], model, self.hit_weights)
with self.assertRaises(MacsypyError) as ctx:
c1.score
self.assertEqual(str(ctx.exception),
"a Cluster contains hit which is neither mandatory nor accessory")
def test_likely_systems_to_tsv(self):
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'unordered'
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)
# test if id is well incremented
gene_name = "gspD"
c_gene_gspd = CoreGene(models_location, gene_name, profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
gene_name = "sctJ"
c_gene_sctj = CoreGene(models_location, gene_name, profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
gene_name = "sctC"
c_gene_sctc = CoreGene(models_location, gene_name, profile_factory)
gene_sctc = ModelGene(c_gene_sctc, model)
model.add_neutral_gene(gene_sctc)
gene_name = "tadZ"
c_gene_tadz = CoreGene(models_location, gene_name, profile_factory)
gene_tadz = ModelGene(c_gene_tadz, model)
model.add_forbidden_gene(gene_tadz)
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", 804, "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_sctc, "hit_3", 805, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(hit_3, gene_sctc, GeneStatus.NEUTRAL)
hit_4 = Hit(c_gene_tadz, "hit_4", 806, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_4 = ValidHit(hit_4, gene_tadz, GeneStatus.FORBIDDEN)
system_1 = LikelySystem(model, [v_hit_1], [v_hit_2], [v_hit_3],
[v_hit_4])
sol_tsv = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Likely Systems found:"""
sol_tsv += "\n\n"
sol_tsv += "\t".join([
"replicon", "hit_id", "gene_name", "hit_pos", "model_fqn",
"sys_id", "sys_wholeness", "hit_gene_ref", "hit_status",
"hit_seq_len", "hit_i_eval", "hit_score", "hit_profile_cov",
"hit_seq_cov", "hit_begin_match", "hit_end_match", "used_in"
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
"replicon_id", "hit_1", "gspD", "1", "foo/T2SS",
"replicon_id_T2SS_1", "1.000", "gspD", "mandatory", "803", "1.0",
"1.000", "1.000", "1.000", "10", "20", ""
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
"replicon_id", "hit_2", "sctJ", "1", "foo/T2SS",
"replicon_id_T2SS_1", "1.000", "sctJ", "accessory", "804", "1.0",
"1.000", "1.000", "1.000", "10", "20", ""
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
"replicon_id", "hit_4", "tadZ", "1", "foo/T2SS",
"replicon_id_T2SS_1", "1.000", "tadZ", "forbidden", "806", "1.0",
"1.000", "1.000", "1.000", "10", "20", ""
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
"replicon_id", "hit_3", "sctC", "1", "foo/T2SS",
"replicon_id_T2SS_1", "1.000", "sctC", "neutral", "805", "1.0",
"1.000", "1.000", "1.000", "10", "20", ""
])
sol_tsv += "\n"
sol_tsv += "\n"
f_out = StringIO()
track_multi_systems_hit = HitSystemTracker([system_1])
likely_systems_to_tsv([system_1], track_multi_systems_hit, f_out)
self.assertMultiLineEqual(sol_tsv, f_out.getvalue())
f_out = StringIO()
likely_systems_to_tsv([], track_multi_systems_hit, f_out)
expected_out = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# No Likely Systems found
"""
self.assertEqual(expected_out, f_out.getvalue())
def test_unnlikely_systems_to_txt(self):
args = argparse.Namespace()
args.sequence_db = self.find_data("base", "test_1.fasta")
args.db_type = 'unordered'
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)
# test if id is well incremented
gene_name = "gspD"
c_gene_gspd = CoreGene(models_location, gene_name, profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
gene_name = "sctJ"
c_gene_sctj = CoreGene(models_location, gene_name, profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model)
model.add_accessory_gene(gene_sctj)
gene_name = "sctC"
c_gene_sctc = CoreGene(models_location, gene_name, profile_factory)
gene_sctc = ModelGene(c_gene_sctc, model)
model.add_neutral_gene(gene_sctc)
gene_name = "tadZ"
c_gene_tadz = CoreGene(models_location, gene_name, profile_factory)
gene_tadz = ModelGene(c_gene_tadz, model)
model.add_forbidden_gene(gene_tadz)
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", 804, "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_sctc, "hit_3", 805, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_3 = ValidHit(hit_3, gene_sctc, GeneStatus.NEUTRAL)
hit_4 = Hit(c_gene_tadz, "hit_4", 806, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
v_hit_4 = ValidHit(hit_4, gene_tadz, GeneStatus.FORBIDDEN)
reason = "why it not a system"
system_1 = UnlikelySystem(model, [v_hit_1], [v_hit_2], [v_hit_3],
[v_hit_4], reason)
exp_txt = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Unlikely Systems found:
This replicon probably not contains a system foo/T2SS:
{reason}
system id = replicon_id_T2SS_1
model = foo/T2SS
replicon = replicon_id
hits = [('hit_1', 'gspD', 1), ('hit_2', 'sctJ', 1), ('hit_3', 'sctC', 1), ('hit_4', 'tadZ', 1)]
wholeness = 1.000
mandatory genes:
\t- gspD: 1 (gspD)
accessory genes:
\t- sctJ: 1 (sctJ)
neutral genes:
\t- sctC: 1 (sctC)
forbidden genes:
\t- tadZ: 1 (tadZ)
Use ordered replicon to have better prediction.
============================================================
"""
f_out = StringIO()
unlikely_systems_to_txt([system_1], f_out)
self.assertMultiLineEqual(exp_txt, f_out.getvalue())
f_out = StringIO()
unlikely_systems_to_txt([], f_out)
expected_out = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# No Unlikely Systems found
"""
self.assertEqual(expected_out, f_out.getvalue())
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 = 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)
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(), 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)
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)
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_flgB)
model_B.add_accessory_gene(gene_tadZ)
h_sctj = Hit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_sctn = Hit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_gspd = Hit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_sctj_flg = Hit(c_gene_sctj_flg, "hit_sctj_flg", 803, "replicon_id",
1, 1.0, 1.0, 1.0, 1.0, 10, 20)
h_flgB = Hit(c_gene_flgB, "hit_flgB", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_tadZ = Hit(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([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.MANDATORY),
ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_A, self.hit_weights)
c2 = Cluster([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(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([
ValidHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ValidHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ValidHit(h_flgB, gene_flgB, 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"
sol = [sys_A, sys_B]
sol_id = '12'
hit_multi_sys_tracker = HitSystemTracker([sys_A, sys_B])
system_serializer = TsvSolutionSerializer()
sol_tsv = '\t'.join([
sol_id, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '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', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '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', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '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', '1.000', '1.500', '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', '1', 'foo/A',
'sys_id_A', '2', '1.000', '1.500', '2', 'sctN', 'mandatory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctj_flg', 'sctJ_FLG', '1', 'foo/B',
'sys_id_B', '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_tadZ', 'tadZ', '1', 'foo/B',
'sys_id_B', '1', '0.750', '2.000', '1', 'tadZ', 'accessory', '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', '1', 'foo/B',
'sys_id_B', '1', '0.750', '2.000', '1', 'flgB', 'accessory', '803',
'1.0', '1.000', '1.000', '1.000', '10', '20', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
ser = system_serializer.serialize(sol, sol_id, hit_multi_sys_tracker)
self.assertEqual(ser, sol_tsv)
def test_rejected_clst_to_txt(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')
args.res_search_dir = "blabla"
cfg = Config(MacsyDefaults(), args)
model_name = 'foo'
models_location = ModelLocation(
path=os.path.join(args.models_dir, model_name))
profile_factory = ProfileFactory(cfg)
model = Model("foo/T2SS", 11)
gene_name = "gspD"
c_gene_gspd = CoreGene(models_location, gene_name, profile_factory)
gene_1 = ModelGene(c_gene_gspd, model)
gene_name = "sctC"
c_gene_sctc = CoreGene(models_location, gene_name, profile_factory)
gene_2 = ModelGene(c_gene_sctc, model)
model.add_mandatory_gene(gene_1)
model.add_accessory_gene(gene_2)
# 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_gspd, "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_sctc, "h20", 10, "replicon_1", 20, 1.0, 20.0, 1.0,
1.0, 10, 20)
v_h20 = ValidHit(h20, gene_2, GeneStatus.ACCESSORY)
h40 = Hit(c_gene_gspd, "h10", 10, "replicon_1", 40, 1.0, 10.0, 1.0,
1.0, 10, 20)
v_h40 = ValidHit(h40, gene_1, GeneStatus.MANDATORY)
h50 = Hit(c_gene_sctc, "h20", 10, "replicon_1", 50, 1.0, 20.0, 1.0,
1.0, 10, 20)
v_h50 = ValidHit(h50, gene_2, GeneStatus.ACCESSORY)
hit_weights = HitWeight(**cfg.hit_weights())
c1 = Cluster([v_h10, v_h20], model, hit_weights)
c2 = Cluster([v_h40, v_h50], model, hit_weights)
r_c = RejectedClusters(model, [c1, c2],
["The reasons to reject this clusters"])
rej_clst_str = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Rejected clusters:
Cluster:
- model = T2SS
- replicon = replicon_1
- hits = (h10, gspD, 10), (h20, sctC, 20)
Cluster:
- model = T2SS
- replicon = replicon_1
- hits = (h10, gspD, 40), (h20, sctC, 50)
These clusters have been rejected because:
\t- The reasons to reject this clusters
============================================================
"""
f_out = StringIO()
rejected_clst_to_txt([r_c], f_out)
self.maxDiff = None
self.assertMultiLineEqual(rej_clst_str, f_out.getvalue())
rej_clst_str = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# No Rejected clusters
"""
f_out = StringIO()
rejected_clst_to_txt([], f_out)
self.assertMultiLineEqual(rej_clst_str, f_out.getvalue())
def test_systems_to_txt(self):
system_str = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# No Systems found
"""
f_out = StringIO()
track_multi_systems_hit = HitSystemTracker([])
systems_to_txt([], track_multi_systems_hit, f_out)
self.assertMultiLineEqual(system_str, f_out.getvalue())
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)
# test if id is well incremented
gene_name = "gspD"
c_gene_gspd = CoreGene(models_location, gene_name, profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model)
model.add_mandatory_gene(gene_gspd)
gene_name = "sctJ"
c_gene_sctj = CoreGene(models_location, gene_name, 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)
system_1 = System(model, [
Cluster([v_hit_1, v_hit_2], model, HitWeight(**cfg.hit_weights()))
], cfg.redundancy_penalty())
system_str = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Systems found:
system id = replicon_id_T2SS_{next(System._id) - 1}
model = foo/T2SS
replicon = replicon_id
clusters = [('hit_1', 'gspD', 1), ('hit_2', 'sctJ', 1)]
occ = 1
wholeness = 1.000
loci nb = 1
score = 1.500
mandatory genes:
\t- gspD: 1 (gspD)
accessory genes:
\t- sctJ: 1 (sctJ)
neutral genes:
============================================================
"""
f_out = StringIO()
track_multi_systems_hit = HitSystemTracker([system_1])
systems_to_txt([system_1], track_multi_systems_hit, f_out)
self.assertMultiLineEqual(system_str, f_out.getvalue())
def test_SystemSerializer_str(self):
model_name = 'foo'
model_location = ModelLocation(
path=os.path.join(self.cfg.models_dir(), 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 = Hit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_sctn = Hit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_gspd = Hit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 1, 1.0, 1.0,
1.0, 1.0, 10, 20)
h_sctj_flg = Hit(c_gene_sctj_flg, "hit_sctj_flg", 803, "replicon_id",
1, 1.0, 1.0, 1.0, 1.0, 10, 20)
h_tadZ = Hit(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([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.MANDATORY),
ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_A, self.hit_weights)
c2 = Cluster([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(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([
ValidHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ValidHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ValidHit(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 _build_systems(cfg, profile_factory):
model_name = 'foo'
model_location = ModelLocation(
path=os.path.join(cfg.models_dir(), model_name))
model_A = Model("foo/A", 10)
model_B = Model("foo/B", 10)
model_C = Model("foo/C", 10)
model_D = Model("foo/D", 10)
model_E = Model("foo/E", 10)
model_F = Model("foo/F", 10)
model_G = Model("foo/G", 10)
model_H = Model("foo/H", 10)
c_gene_sctn_flg = CoreGene(model_location, "sctN_FLG", profile_factory)
gene_sctn_flg = ModelGene(c_gene_sctn_flg, model_B)
c_gene_sctj_flg = CoreGene(model_location, "sctJ_FLG", profile_factory)
gene_sctj_flg = ModelGene(c_gene_sctj_flg, model_B)
c_gene_flgB = CoreGene(model_location, "flgB", profile_factory)
gene_flgB = ModelGene(c_gene_flgB, model_B)
c_gene_tadZ = CoreGene(model_location, "tadZ", profile_factory)
gene_tadZ = ModelGene(c_gene_tadZ, model_B)
c_gene_sctn = CoreGene(model_location, "sctN", 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", 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", 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", 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_flgB)
model_B.add_accessory_gene(gene_tadZ)
model_C.add_mandatory_gene(gene_sctn_flg)
model_C.add_mandatory_gene(gene_sctj_flg)
model_C.add_mandatory_gene(gene_flgB)
model_C.add_accessory_gene(gene_tadZ)
model_C.add_accessory_gene(gene_gspd)
model_D.add_mandatory_gene(gene_abc)
model_D.add_accessory_gene(gene_sctn)
model_E.add_accessory_gene(gene_gspd)
model_F.add_mandatory_gene(gene_abc)
# idem as C
model_G.add_mandatory_gene(gene_sctn_flg)
model_G.add_mandatory_gene(gene_sctj_flg)
model_G.add_mandatory_gene(gene_flgB)
model_G.add_accessory_gene(gene_tadZ)
model_G.add_accessory_gene(gene_gspd)
# idem as D
model_H.add_mandatory_gene(gene_abc)
model_H.add_accessory_gene(gene_sctn)
h_sctj = Hit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0, 1.0, 1.0,
1.0, 10, 20)
h_sctn = Hit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 2, 1.0, 1.0, 1.0,
1.0, 10, 20)
h_gspd = Hit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 3, 1.0, 1.0, 1.0,
1.0, 10, 20)
h_sctj_flg = Hit(c_gene_sctj_flg, "hit_sctj_flg", 803, "replicon_id", 4,
1.0, 1.0, 1.0, 1.0, 10, 20)
h_flgB = Hit(c_gene_flgB, "hit_flgB", 803, "replicon_id", 5, 1.0, 1.0, 1.0,
1.0, 10, 20)
h_tadZ = Hit(c_gene_tadZ, "hit_tadZ", 803, "replicon_id", 6, 1.0, 1.0, 1.0,
1.0, 10, 20)
h_abc = Hit(c_gene_abc, "hit_abc", 803, "replicon_id", 7, 1.0, 1.0, 1.0,
1.0, 10, 20)
model_A._min_mandatory_genes_required = 2
model_A._min_genes_required = 2
hit_weights = HitWeight(**cfg.hit_weights())
c1 = Cluster([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.MANDATORY),
ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_A, hit_weights)
c2 = Cluster([
ValidHit(h_sctj, gene_sctj, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.MANDATORY)
], model_A, hit_weights)
model_B._min_mandatory_genes_required = 1
model_B._min_genes_required = 2
c3 = Cluster([
ValidHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ValidHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ValidHit(h_flgB, gene_flgB, GeneStatus.ACCESSORY)
], model_B, hit_weights)
model_C._min_mandatory_genes_required = 1
model_C._min_genes_required = 2
c4 = Cluster([
ValidHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY),
ValidHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY),
ValidHit(h_flgB, gene_flgB, GeneStatus.MANDATORY),
ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
], model_C, hit_weights)
model_D._min_mandatory_genes_required = 1
model_D._min_genes_required = 1
c5 = Cluster([
ValidHit(h_abc, gene_abc, GeneStatus.MANDATORY),
ValidHit(h_sctn, gene_sctn, GeneStatus.ACCESSORY)
], model_D, hit_weights)
model_E._min_mandatory_genes_required = 0
model_E._min_genes_required = 1
c6 = Cluster([ValidHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)], model_E,
hit_weights)
model_F._min_mandatory_genes_required = 1
model_F._min_genes_required = 1
c7 = Cluster([ValidHit(h_abc, gene_abc, GeneStatus.MANDATORY)], model_F,
hit_weights)
systems = {}
systems['A'] = System(model_A, [c1, c2],
cfg.redundancy_penalty()) # 5 hits
# we need to tweek the replicon_id to have stable ressults
# whatever the number of tests ran
# or the tests order
systems['A'].id = "replicon_id_A"
systems['B'] = System(model_B, [c3], cfg.redundancy_penalty()) # 3 hits
systems['B'].id = "replicon_id_B"
systems['C'] = System(model_C, [c4], cfg.redundancy_penalty()) # 4 hits
systems['C'].id = "replicon_id_C"
systems['D'] = System(model_D, [c5], cfg.redundancy_penalty()) # 2 hits
systems['D'].id = "replicon_id_D"
systems['E'] = System(model_E, [c6], cfg.redundancy_penalty()) # 1 hit
systems['E'].id = "replicon_id_E"
systems['F'] = System(model_F, [c7], cfg.redundancy_penalty()) # 1 hit
systems['F'].id = "replicon_id_F"
systems['G'] = System(model_G, [c4], cfg.redundancy_penalty()) # 4 hits
systems['G'].id = "replicon_id_G"
systems['H'] = System(model_H, [c5], cfg.redundancy_penalty()) # 2 hits
systems['H'].id = "replicon_id_H"
return systems