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_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 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_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_likely_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)
system_1 = LikelySystem(model, [v_hit_1], [v_hit_2], [v_hit_3],
[v_hit_4])
system_str = f"""# macsyfinder {macsypy.__version__}
# {' '.join(sys.argv)}
# Systems found:
This replicon contains genetic materials needed for system foo/T2SS
WARNING there quorum is reached but there is also some forbidden genes.
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()
track_multi_systems_hit = HitSystemTracker([system_1])
likely_systems_to_txt([system_1], track_multi_systems_hit, f_out)
self.assertMultiLineEqual(system_str, f_out.getvalue())
f_out = StringIO()
likely_systems_to_txt([], 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_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_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_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 main(args=None, loglevel=None):
"""
main entry point to MacSyFinder do some check before to launch :func:`main_search_systems` which is
the real function that perform a search
:param args: the arguments passed on the command line without the program name
:type args: List of string
:param loglevel: the output verbosity
:type loglevel: a positive int or a string among 'DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL'
"""
args = sys.argv[1:] if args is None else args
parser, parsed_args = parse_args(args)
defaults = MacsyDefaults()
config = Config(defaults, parsed_args)
###########################
# creation of working dir
###########################
working_dir = config.working_dir()
if not os.path.exists(working_dir):
os.makedirs(working_dir)
else:
if os.path.isdir(working_dir):
if os.listdir(working_dir):
raise ValueError(
f"'{working_dir}' already exists and is not a empty")
else:
raise ValueError(
f"'{working_dir}' already exists and is not a directory")
################
# init loggers #
################
macsypy.init_logger(log_file=os.path.join(config.working_dir(),
config.log_file()),
out=not config.mute())
if not loglevel:
# logs are specify from args options
macsypy.logger_set_level(level=config.log_level())
else:
# used by unit tests to mute or unmute logs
macsypy.logger_set_level(level=loglevel)
logger = logging.getLogger('macsypy.macsyfinder')
if parsed_args.list_models:
print(list_models(parsed_args), file=sys.stdout)
sys.exit(0)
else:
if not parsed_args.previous_run and not parsed_args.models:
parser.print_help()
print()
sys.tracebacklimit = 0
raise OptionError(
"argument --models or --previous-run is required.")
elif not parsed_args.previous_run and not parsed_args.sequence_db:
parser.print_help()
print()
sys.tracebacklimit = 0
raise OptionError(
"argument --sequence-db or --previous-run is required.")
elif not parsed_args.previous_run and not parsed_args.db_type:
parser.print_help()
print()
sys.tracebacklimit = 0
raise OptionError(
"argument --db-type or --previous-run is required.")
_log.info(f"command used: {' '.join(sys.argv)}")
models = ModelBank()
genes = GeneBank()
profile_factory = ProfileFactory(config)
macsypy.hit.hit_weight = macsypy.hit.HitWeight(itself=3,
exchangeable=.75,
mandatory=2,
accessory=.25,
neutral=1.5)
logger.info("\n{:#^70}".format(" Searching systems "))
all_systems, rejected_clusters = search_systems(
config, models, genes, profile_factory, logger)
track_multi_systems_hit = HitSystemTracker(all_systems)
if config.db_type() in ('gembase', 'ordered_replicon'):
#############################
# Ordered/Gembase replicons #
#############################
###########################
# select the best systems #
###########################
logger.info("\n{:#^70}".format(" Computing best solutions "))
best_solutions = []
one_best_solution = []
# group systems found by replicon
# before to search best system combination
import time
for rep_name, syst_group in itertools.groupby(
all_systems, key=lambda s: s.replicon_name):
syst_group = list(syst_group)
logger.info(
f"Computing best solutions for {rep_name} (nb of systems {len(syst_group)})"
)
t0 = time.time()
best_sol_4_1_replicon, score = find_best_solutions(syst_group)
t1 = time.time()
logger.info(
f"It took {t1 - t0:.2f}sec to find best solution ({score:.2f}) for replicon {rep_name}"
)
# if several solutions are equivalent same number of system and score is same
# store all equivalent solution in best_solution => all_best_systems
# pick one in one_best_solution => best_systems
best_solutions.extend(best_sol_4_1_replicon)
one_best_solution.append(best_sol_4_1_replicon[0])
##############################
# Write the results in files #
##############################
logger.info("\n{:#^70}".format(" Writing down results "))
system_filename = os.path.join(config.working_dir(),
"all_systems.txt")
tsv_filename = os.path.join(config.working_dir(),
"all_systems.tsv")
with open(system_filename, "w") as sys_file:
systems_to_txt(all_systems, track_multi_systems_hit, sys_file)
with open(tsv_filename, "w") as tsv_file:
systems_to_tsv(all_systems, track_multi_systems_hit, tsv_file)
cluster_filename = os.path.join(config.working_dir(),
"rejected_clusters.txt")
with open(cluster_filename, "w") as clst_file:
rejected_clusters.sort(key=lambda clst: (
clst.replicon_name, clst.model, clst.hits))
rejected_clst_to_txt(rejected_clusters, clst_file)
if not (all_systems or rejected_clusters):
logger.info("No Systems found in this dataset.")
tsv_filename = os.path.join(config.working_dir(),
"all_best_solutions.tsv")
with open(tsv_filename, "w") as tsv_file:
solutions_to_tsv(best_solutions, track_multi_systems_hit,
tsv_file)
tsv_filename = os.path.join(config.working_dir(),
"best_solution.tsv")
with open(tsv_filename, "w") as tsv_file:
# flattern the list and sort it
one_best_solution = [
syst for sol in one_best_solution for syst in sol
]
one_best_solution.sort(
key=lambda syst: (syst.replicon_name, syst.position[0],
syst.model.fqn, -syst.score))
systems_to_tsv(one_best_solution, track_multi_systems_hit,
tsv_file)
else:
#######################
# Unordered replicons #
#######################
##############################
# Write the results in files #
##############################
logger.info("\n{:#^70}".format(" Writing down results "))
system_filename = os.path.join(config.working_dir(),
"all_systems.txt")
with open(system_filename, "w") as sys_file:
likely_systems_to_txt(all_systems, track_multi_systems_hit,
sys_file)
# forbidden = [s for s in all_systems if s.forbidden_occ]
# system_filename = os.path.join(config.working_dir(), "forbidden_components.tsv")
# with open(system_filename, "w") as sys_file:
# likely_systems_to_tsv(forbidden, track_multi_systems_hit, sys_file)
system_filename = os.path.join(config.working_dir(),
"all_systems.tsv")
with open(system_filename, "w") as sys_file:
likely_systems_to_tsv(all_systems, track_multi_systems_hit,
sys_file)
cluster_filename = os.path.join(config.working_dir(),
"uncomplete_systems.txt")
with open(cluster_filename, "w") as clst_file:
unlikely_systems_to_txt(rejected_clusters, clst_file)
if not (all_systems or rejected_clusters):
logger.info("No Systems found in this dataset.")
logger.info("END")
def test_SolutionSerializer_tsv(self):
model_name = 'foo'
model_location = ModelLocation(
path=os.path.join(self.cfg.models_dir()[0], model_name))
###########
# Model B #
###########
model_B = Model("foo/B", 10)
c_gene_sctn_flg = CoreGene(model_location, "sctN_FLG",
self.profile_factory)
gene_sctn_flg = ModelGene(c_gene_sctn_flg, model_B)
c_gene_sctj_flg = CoreGene(model_location, "sctJ_FLG",
self.profile_factory)
gene_sctj_flg = ModelGene(c_gene_sctj_flg, model_B)
c_gene_flgB = CoreGene(model_location, "flgB", self.profile_factory)
gene_flgB = ModelGene(c_gene_flgB, model_B)
c_gene_tadZ = CoreGene(model_location, "tadZ", self.profile_factory)
gene_tadZ = ModelGene(c_gene_tadZ, model_B)
model_B.add_mandatory_gene(gene_sctn_flg)
model_B.add_mandatory_gene(gene_sctj_flg)
model_B.add_accessory_gene(gene_flgB)
model_B.add_accessory_gene(gene_tadZ)
###########
# Model A #
###########
model_A = Model("foo/A", 10)
c_gene_sctn = CoreGene(model_location, "sctN", self.profile_factory)
gene_sctn = ModelGene(c_gene_sctn, model_A)
gene_sctn_hom = Exchangeable(c_gene_sctn_flg, gene_sctn)
gene_sctn.add_exchangeable(gene_sctn_hom)
c_gene_sctj = CoreGene(model_location, "sctJ", self.profile_factory)
gene_sctj = ModelGene(c_gene_sctj, model_A)
gene_sctj_an = Exchangeable(c_gene_sctj_flg, gene_sctj)
gene_sctj.add_exchangeable(gene_sctj_an)
c_gene_gspd = CoreGene(model_location, "gspD", self.profile_factory)
gene_gspd = ModelGene(c_gene_gspd, model_A)
gene_gspd_an = Exchangeable(c_gene_flgB, gene_gspd)
gene_gspd.add_exchangeable(gene_gspd_an)
c_gene_abc = CoreGene(model_location, "abc", self.profile_factory)
gene_abc = ModelGene(c_gene_abc, model_A, loner=True)
gene_abc_ho = Exchangeable(c_gene_tadZ, gene_abc)
gene_abc.add_exchangeable(gene_abc_ho)
model_A.add_mandatory_gene(gene_sctn)
model_A.add_mandatory_gene(gene_sctj)
model_A.add_accessory_gene(gene_gspd)
model_A.add_accessory_gene(gene_abc)
# CoreHit(gene, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
# pos score
h_sctj = CoreHit(c_gene_sctj, "hit_sctj", 803, "replicon_id", 1, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_sctj = ModelHit(h_sctj, gene_sctj, GeneStatus.MANDATORY)
h_sctn = CoreHit(c_gene_sctn, "hit_sctn", 803, "replicon_id", 2, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_sctn = ModelHit(h_sctn, gene_sctn, GeneStatus.MANDATORY)
h_gspd = CoreHit(c_gene_gspd, "hit_gspd", 803, "replicon_id", 3, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_gspd = ModelHit(h_gspd, gene_gspd, GeneStatus.ACCESSORY)
h_sctj_flg = CoreHit(c_gene_sctj_flg, "hit_sctj_flg", 803,
"replicon_id", 10, 1.0, 1.0, 1.0, 1.0, 10, 20)
h_flgB = CoreHit(c_gene_flgB, "hit_flgB", 803, "replicon_id", 11, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_abc = CoreHit(c_gene_abc, "hit_abc", 803, "replicon_id", 20, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_abc2 = CoreHit(c_gene_abc, "hit_abc2", 803, "replicon_id", 50, 1.0,
1.0, 1.0, 1.0, 10, 20)
h_tadZ = CoreHit(c_gene_tadZ, "hit_tadZ", 803, "replicon_id", 40, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_sctj_flg = ModelHit(h_sctj_flg, gene_sctj_flg, GeneStatus.MANDATORY)
mh_flgB = ModelHit(h_flgB, gene_flgB, GeneStatus.ACCESSORY)
mh_abc = ModelHit(h_abc, gene_abc, GeneStatus.ACCESSORY)
mh_abc2 = ModelHit(h_abc2, gene_abc, GeneStatus.ACCESSORY)
mh_tadZ = ModelHit(h_tadZ, gene_tadZ, GeneStatus.ACCESSORY)
model_A._min_mandatory_genes_required = 2
model_A._min_genes_required = 2
c1 = Cluster([mh_sctj, mh_sctn, mh_gspd], model_A, self.hit_weights)
c2 = Cluster([mh_sctj, mh_sctn], model_A, self.hit_weights)
c3 = Cluster([
Loner(h_abc,
gene_ref=gene_abc,
gene_status=GeneStatus.ACCESSORY,
counterpart=[mh_abc2])
], model_A, self.hit_weights)
model_B._min_mandatory_genes_required = 1
model_B._min_genes_required = 2
c5 = Cluster([mh_sctj_flg, mh_tadZ, mh_flgB], model_B,
self.hit_weights)
sys_A = System(model_A, [c1, c2, c3], self.cfg.redundancy_penalty())
# score = 2.5, 2 , 0.35 = 4.85 - (2 * 1.5) = 1.85
sys_A.id = "sys_id_A"
sys_B = System(model_B, [c5], self.cfg.redundancy_penalty())
# score = 2.0
sys_B.id = "sys_id_B"
sol = Solution([sys_A, sys_B])
sol_id = '12'
hit_multi_sys_tracker = HitSystemTracker([sys_A, sys_B])
sol_serializer = TsvSolutionSerializer()
sol_tsv = '\t'.join([
sol_id, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '1', '1.000', '1.850', '2', 'sctJ', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctn', 'sctN', '2', 'foo/A',
'sys_id_A', '2', '1', '1.000', '1.850', '2', 'sctN', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_gspd', 'gspD', '3', 'foo/A',
'sys_id_A', '2', '1', '1.000', '1.850', '2', 'gspD', 'accessory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctj', 'sctJ', '1', 'foo/A',
'sys_id_A', '2', '2', '1.000', '1.850', '2', 'sctJ', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctn', 'sctN', '2', 'foo/A',
'sys_id_A', '2', '2', '1.000', '1.850', '2', 'sctN', 'mandatory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_abc', 'abc', '20', 'foo/A', 'sys_id_A',
'2', '-1', '1.000', '1.850', '2', 'abc', 'accessory', '803', '1.0',
'1.000', '1.000', '1.000', '10', '20', 'hit_abc2', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_sctj_flg', 'sctJ_FLG', '10', 'foo/B',
'sys_id_B', '1', '1', '0.750', '2.000', '1', 'sctJ_FLG',
'mandatory', '803', '1.0', '1.000', '1.000', '1.000', '10', '20',
'', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_flgB', 'flgB', '11', 'foo/B',
'sys_id_B', '1', '1', '0.750', '2.000', '1', 'flgB', 'accessory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += '\t'.join([
sol_id, 'replicon_id', 'hit_tadZ', 'tadZ', '40', 'foo/B',
'sys_id_B', '1', '1', '0.750', '2.000', '1', 'tadZ', 'accessory',
'803', '1.0', '1.000', '1.000', '1.000', '10', '20', '', ''
])
sol_tsv += "\n"
sol_tsv += "\n"
ser = sol_serializer.serialize(sol, sol_id, hit_multi_sys_tracker)
self.maxDiff = None
self.assertEqual(ser, sol_tsv)
def test_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, loner=True)
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)
#CoreHit(gene, hit_id, hit_seq_length, replicon_name, position, i_eval, score,
# profile_coverage, sequence_coverage, begin_match, end_match
# pos score
ch_gspd = CoreHit(c_gene_gspd, "h_gspd", 803, "replicon_id", 10, 1.0,
1.0, 1.0, 1.0, 10, 20)
mh_gspd = ModelHit(ch_gspd,
gene_ref=gene_gspd,
gene_status=GeneStatus.MANDATORY)
ch_sctj = CoreHit(c_gene_sctj, "h_sctj", 803, "replicon_id", 20, 1.0,
1.0, 1.0, 1.0, 20, 30)
mh_sctj = ModelHit(ch_sctj,
gene_ref=gene_sctj,
gene_status=GeneStatus.ACCESSORY)
ch_sctn_flg = CoreHit(c_gene_sctn_flg, "h_sctn_flg", 803,
"replicon_id", 40, 1.0, 1.0, 1.0, 1.0, 30, 40)
mh_sctn_flg = ModelHit(ch_sctn_flg,
gene_ref=gene_sctn_flg,
gene_status=GeneStatus.ACCESSORY)
ch_sctn = CoreHit(c_gene_sctn, "h_sctn", 803, "replicon_id", 80, 1.0,
1.0, 1.0, 1.0, 30, 40)
mh_sctn = Loner(ch_sctn,
gene_ref=gene_sctn,
gene_status=GeneStatus.ACCESSORY,
counterpart=[mh_sctn_flg])
c1 = Cluster([mh_gspd, mh_sctj], model, self.hit_weights)
c2 = Cluster([mh_sctn], model, self.hit_weights)
sys_multi_loci = System(model, [c1, c2], self.cfg.redundancy_penalty())
# score 1.5 .35 = 1.85
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", "1.000", "1.850", "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", "1.000", "1.850", "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", "sctN", "80", "foo/T2SS",
sys_multi_loci.id, "1", "-1", "1.000", "1.850", "1", "sctN",
"accessory", "803", "1.0", "1.000", "1.000", "1.000", "30", "40",
"h_sctn_flg", ""
])
sys_tsv += "\n"
self.maxDiff = None
self.assertEqual(
sys_tsv,
system_serializer.serialize(sys_multi_loci, hit_multi_sys_tracker))