def test_iter(self):
# test Gembase Draft
seq_name = 'ACBA.0917.00019'
ext = '.fna'
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
db = GembaseDB(replicon, cfg)
idx = SeqIO.index(self.find_data(
os.path.join('Gembase', 'Proteins', seq_name + '.prt')),
'fasta',
alphabet=Seq.IUPAC.extended_protein)
specie, date, strain, contig = replicon.id.split('.')
pattern = '{}\.{}\.{}\.\w?{}'.format(specie, date, strain, contig)
self.assertListEqual(sorted([i for i in idx if re.match(pattern, i)]),
sorted([i for i in db]))
# test Gembase Complet
seq_name = 'ESCO001.C.00001.C001'
ext = '.fst'
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
with self.catch_log():
db = GembaseDB(replicon, cfg)
idx = SeqIO.index(self.find_data(
os.path.join('Gembase', 'Proteins', seq_name + '.prt')),
'fasta',
alphabet=Seq.IUPAC.extended_protein)
specie, date, strain, contig = replicon.id.split('.')
pattern = '{}\.{}\.{}\.\w?{}'.format(specie, date, strain, contig)
seqid_from_gembase_protfile = set(
[i for i in idx if re.match(pattern, i)])
seqid_from_if = set([i for i in db])
non_common_seq = seqid_from_gembase_protfile ^ seqid_from_if
# in Gembase complete the annotation from lstinfo provided from genbank
# it appear some times that some CDS are not translate in proteins
# So in data I have 3 genes from LSTINFO are not in .prt file
diff = {
'ESCO001.C.00001.C001_03974', 'ESCO001.C.00001.C001_01509',
'ESCO001.C.00001.C001_04162'
}
self.assertSetEqual(non_common_seq, diff)
def test_find_gembase_file_basename_file_not_in_gembase(self):
"""
test if find_gembase_file_basename get the the right basename
for files not located in gembase and file name is the output of split operation
a file containing one contig
a file representing a chunk
"""
gembase_path = self.find_data('Gembase')
file_names = {
'ACBA.0917.00019':
self.find_data(
os.path.join('Replicons', 'ACBA.0917.00019.0001.fst')),
'ESCO001.C.00001.C001.fst':
os.path.join(self.tmp_dir, 'ESCO001.C.00001.C001_chunk_1.fst')
}
shutil.copyfile(
os.path.join(gembase_path, 'Replicons',
'ESCO001.C.00001.C001.fst'),
file_names['ESCO001.C.00001.C001.fst'])
for base_file_name, replicon_path in file_names.items():
self.args.replicon = replicon_path
self.args.gembase_path = gembase_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg, gembase_path=gembase_path)
self.assertTrue(
db._find_gembase_file_basename(gembase_path, replicon_path),
base_file_name)
replicon_path = self.find_data(
os.path.join('Replicons', 'acba.007.p01.13.fst'))
self.args.replicon = replicon_path
self.args.gembase_path = gembase_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.assertRaises(FileNotFoundError) as ctx:
with self.catch_log():
GembaseDB(replicon, cfg, gembase_path=gembase_path)
self.assertEqual(
str(ctx.exception),
'cannot find lst file matching {} sequence'.format(replicon_path))
def test_gembase_draft_parser(self):
replicon_name = 'ACBA.0917.00019'
replicon_id = 'ACBA.0917.00019.0001'
lst_path = self.find_data(
os.path.join('Gembase', 'LSTINFO', replicon_name + '.lst'))
prots_info = GembaseDB.gembase_draft_parser(lst_path, replicon_id)
columns = [
'start', 'end', 'strand', 'type', 'seq_id', 'gene_name',
'description'
]
self.assertListEqual(list(prots_info.columns), columns)
self.assertEqual(prots_info.shape, (3870, len(columns)))
first_row = [
266, 1480, 'C', 'CDS', 'ACBA.0917.00019.b0001_00001', 'tyrS',
'| Tyrosine--tRNA ligase | 6.1.1.1 | similar to AA sequence:UniProtKB:P41256'
]
recieved_first_row = prots_info.iloc[0].values.tolist()
self.assertListEqual(first_row, recieved_first_row)
last_row = [
4043755, 4044354, 'C', 'CDS', 'ACBA.0917.00019.i0001_03957',
'yfcG_3',
'| Disulfide-bond oxidoreductase YfcG | 1.8.4.- | similar to AA sequence:UniProtKB:P77526'
]
recieved_last_row = prots_info.iloc[len(prots_info) -
1].values.tolist()
self.assertListEqual(last_row, recieved_last_row)
def test_make_protfile(self):
file_name = (('ACBA.0917.00019', '.fna', 3870),
('ESCO001.C.00001.C001', '.fst', 3870))
for seq_name, ext, seq_nb in file_name:
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg)
for seq_nb, seqs in enumerate(
zip(
read_multi_prot_fasta(
self.find_data(
os.path.join('Gembase', 'Proteins',
seq_name + '.prt'))),
read_multi_prot_fasta(db.protfile)), 1):
expected, test = seqs
self.assertEqual(expected.id, test.id)
self.assertEqual(seq_nb, seq_nb)
def test_getitem(self):
file_name = (('ACBA.0917.00019', '.fna'), ('ESCO001.C.00001.C001',
'.fst'))
for seq_name, ext in file_name:
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg)
exp = read_multi_prot_fasta(
self.find_data(
os.path.join('Gembase', 'Proteins', seq_name + '.prt')))
specie, date, strain, contig = replicon.id.split('.')
pattern = '{}\.{}\.{}\.\w?{}'.format(specie, date, strain, contig)
for prot_expected in exp:
if re.match(pattern, prot_expected.id):
prot_received = db[prot_expected.id]
self.assertEqual(prot_received.id, prot_expected.id)
self.assertEqual(prot_received.seq, prot_expected.seq)
with self.assertRaises(KeyError) as ctx:
db['nimport_naoik']
self.assertEqual(str(ctx.exception), "'nimport_naoik'")
def test_gembase_complete_parser(self):
replicon_id = 'ESCO001.C.00001.C001'
lst_path = self.find_data(
os.path.join('Gembase', 'LSTINFO', replicon_id + '.lst'))
prots_info = GembaseDB.gembase_complete_parser(lst_path, replicon_id)
columns = [
'start', 'end', 'strand', 'type', 'seq_id', 'valid', 'gene_name',
'description'
]
self.assertListEqual(list(prots_info.columns), columns)
self.assertEqual(prots_info.shape, (4139, len(columns)))
first_row = [
190, 255, 'D', 'CDS', 'ESCO001.C.00001.C001_00001', 'Valid',
'thrL',
'@b0001@NP_414542.1@ b0001 1 190 255 | leader; Amino acid biosynthesis:'
' Threonine thr operon leader peptide | ..'
]
recieved_first_row = prots_info.iloc[0].values.tolist()
self.assertListEqual(first_row, recieved_first_row)
last_row = [
4640942, 4641628, 'D', 'CDS', 'ESCO001.C.00001.C001_04495',
'Valid', 'yjtD',
'@b4403@NP_418820.1@ b4403 1 4640942 4641628 | putative methyltransferase | ..'
]
recieved_last_row = prots_info.iloc[len(prots_info) -
1].values.tolist()
self.assertListEqual(last_row, recieved_last_row)
def test_gembase_sniffer(self):
file_names = (('ACBA.0917.00019', 'Draft'), ('ESCO001.C.00001.C001',
'Complet'))
for file_name, gem_type in file_names:
lst_path = self.find_data(
os.path.join('Gembase', 'LSTINFO', file_name + '.lst'))
type_recieved = GembaseDB.gembase_sniffer(lst_path)
self.assertEqual(type_recieved, gem_type)
def test_find_gembase_file_basename_file_not_in_gembase(self):
"""
test if find_gembase_file_basename get the the right basename
for files not located in gembase and file name is the output of split operation
a file containing one contig
a file representing a chunk
"""
gembase_path = self.find_data('Gembase')
file_names = {'ACBA.0917.00019': self.find_data(os.path.join('Replicons', 'ACBA.0917.00019.0001.fst')),
'ESCO001.C.00001.C001.fst': os.path.join(self.tmp_dir, 'ESCO001.C.00001.C001_chunk_1.fst')
}
shutil.copyfile(os.path.join(gembase_path, 'Replicons', 'ESCO001.C.00001.C001.fst'),
file_names['ESCO001.C.00001.C001.fst'])
for base_file_name, replicon_path in file_names.items():
self.args.replicon = replicon_path
self.args.gembase_path = gembase_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg, gembase_path=gembase_path)
self.assertTrue(db._find_gembase_file_basename(gembase_path, replicon_path),
base_file_name)
replicon_path = self.find_data(os.path.join('Replicons', 'acba.007.p01.13.fst'))
self.args.replicon = replicon_path
self.args.gembase_path = gembase_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.assertRaises(FileNotFoundError) as ctx:
with self.catch_log():
GembaseDB(replicon, cfg, gembase_path=gembase_path)
self.assertEqual(str(ctx.exception),
'cannot find lst file matching {} sequence'.format(replicon_path))
def test_find_gembase_file_basename(self):
"""
test if find_gembase_file_basename get the the right basename
for files in gembase
"""
gembase_path = self.find_data('Gembase')
file_names = ('ACBA.0917.00019.fna', 'ESCO001.C.00001.C001.fst')
for file_name in file_names:
replicon_path = self.find_data(os.path.join('Gembase', 'Replicons', file_name))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg)
self.assertTrue(db._find_gembase_file_basename(gembase_path, replicon_path),
os.path.splitext(file_name)[0])
def test_get_description(self):
# SeqDesc(id, strand, strat, stop)
file_name = {('ACBA.0917.00019', '.fna'):
{'ACBA.0917.00019.b0001_00001': SeqDesc('ACBA.0917.00019.b0001_00001', -1, 266, 1480),
'ACBA.0917.00019.i0001_03957': SeqDesc('ACBA.0917.00019.i0001_03957', -1, 4043755, 4044354)},
}
for seq_name, ext in file_name:
replicon_path = self.find_data(os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
db = GembaseDB(replicon, cfg)
descriptions = file_name[(seq_name, ext)]
for seq_id, desc in descriptions.items():
self.assertEqual(desc, db.get_description(seq_id))
with self.assertRaises(IntegronError) as ctx:
db.get_description('nimport_naoik')
self.assertEqual(str(ctx.exception), "'nimport_naoik' is not a valid Gembase protein identifier.")
with self.assertRaises(KeyError) as ctx:
db.get_description('FOO.BAR.00019.i0001_03924')
self.assertEqual(str(ctx.exception), "'FOO.BAR.00019.i0001_03924'")
def test_find_gembase_file_basename(self):
"""
test if find_gembase_file_basename get the the right basename
for files in gembase
"""
gembase_path = self.find_data('Gembase')
file_names = ('ACBA.0917.00019.fna', 'ESCO001.C.00001.C001.fst')
for file_name in file_names:
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', file_name))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg)
self.assertTrue(
db._find_gembase_file_basename(gembase_path, replicon_path),
os.path.splitext(file_name)[0])
def test_ProteinDB(self):
# From Gembase Draft , Gembase Complete
file_names = ('ACBA.0917.00019.fna', 'ESCO001.C.00001.C001.fst')
for file_name in file_names:
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', file_name))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg)
self.assertTrue(db.replicon.id, replicon.id)
def test_read_multi(self):
"""
Test reading hmm results when there are multiple hits: 2 hits on the same protein: keep
only the one with the best evalue. 2 hits on 2 different proteins: keep the 2 proteins.
"""
replicon_id = 'ACBA.0917.00019'
contig_id = 'ACBA.0917.00019.0001'
result_dir_expected = self.find_data(
"Results_Integron_Finder_{}.gembase".format(replicon_id))
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', replicon_id + '.fna'))
prot_file = os.path.join(result_dir_expected,
"tmp_{}".format(contig_id),
contig_id + '.prt')
args = argparse.Namespace()
args.gembase = True
args.replicon = replicon_path
cfg = Config(args)
sequences_db = read_multi_prot_fasta(replicon_path)
replicon = next(sequences_db)
prot_db = GembaseDB(replicon, cfg, prot_file=prot_file)
infile = self.find_data(
os.path.join('fictive_results',
"{}_intI_multi.res".format(contig_id)))
df = read_hmm(contig_id, prot_db, infile, cfg)
exp = pd.DataFrame(data={
"Accession_number": [contig_id] * 2,
"query_name": ["Phage_integrase"] * 2,
"ID_query": ["PF00589.16"] * 2,
"ID_prot":
["ACBA.0917.00019.i0001_00298", "ACBA.0917.00019.i0001_00338"],
"strand": [-1, -1],
"pos_beg": [311597, 350328],
"pos_end": [312631, 351248],
"evalue": [5.5e-66, 3.4e-51]
},
index=[0, 1])
exp = exp[[
"Accession_number", "query_name", "ID_query", "ID_prot", "strand",
"pos_beg", "pos_end", "evalue"
]]
pdt.assert_frame_equal(df, exp)
def test_protfile(self):
file_name = (('ACBA.0917.00019', '.fna'), ('ESCO001.C.00001.C001',
'.fst'))
for seq_name, ext in file_name:
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
with self.catch_log():
db = GembaseDB(replicon, cfg)
self.assertEqual(
os.path.join(cfg.tmp_dir(replicon.id), replicon.id + '.prt'),
db.protfile)
def test_gembase_complete_parser(self):
replicon_id = 'ESCO001.C.00001.C001'
lst_path = self.find_data(os.path.join('Gembase', 'LSTINFO', replicon_id + '.lst'))
prots_info = GembaseDB.gembase_complete_parser(lst_path, replicon_id)
columns = ['start', 'end', 'strand', 'type', 'seq_id', 'valid', 'gene_name', 'description']
self.assertListEqual(list(prots_info.columns), columns)
self.assertEqual(prots_info.shape, (4139, len(columns)))
first_row = [190, 255, 'D', 'CDS', 'ESCO001.C.00001.C001_00001', 'Valid', 'thrL',
'@b0001@NP_414542.1@ b0001 1 190 255 | leader; Amino acid biosynthesis:'
' Threonine thr operon leader peptide | ..']
recieved_first_row = prots_info.iloc[0].values.tolist()
self.assertListEqual(first_row, recieved_first_row)
last_row = [4640942, 4641628, 'D', 'CDS', 'ESCO001.C.00001.C001_04495', 'Valid', 'yjtD',
'@b4403@NP_418820.1@ b4403 1 4640942 4641628 | putative methyltransferase | ..']
recieved_last_row = prots_info.iloc[len(prots_info) - 1].values.tolist()
self.assertListEqual(last_row, recieved_last_row)
def test_gembase_draft_parser(self):
replicon_name = 'ACBA.0917.00019'
replicon_id = 'ACBA.0917.00019.0001'
lst_path = self.find_data(os.path.join('Gembase', 'LSTINFO', replicon_name + '.lst'))
prots_info = GembaseDB.gembase_draft_parser(lst_path, replicon_id)
columns = ['start', 'end', 'strand', 'type', 'seq_id', 'gene_name', 'description']
self.assertListEqual(list(prots_info.columns), columns)
self.assertEqual(prots_info.shape, (3870, len(columns)))
first_row = [266, 1480, 'C', 'CDS', 'ACBA.0917.00019.b0001_00001', 'tyrS',
'| Tyrosine--tRNA ligase | 6.1.1.1 | similar to AA sequence:UniProtKB:P41256']
recieved_first_row = prots_info.iloc[0].values.tolist()
self.assertListEqual(first_row, recieved_first_row)
last_row = [4043755, 4044354, 'C', 'CDS', 'ACBA.0917.00019.i0001_03957', 'yfcG_3',
'| Disulfide-bond oxidoreductase YfcG | 1.8.4.- | similar to AA sequence:UniProtKB:P77526']
recieved_last_row = prots_info.iloc[len(prots_info) - 1].values.tolist()
self.assertListEqual(last_row, recieved_last_row)
def test_read_hmm_gembase(self):
"""
Test that the hmm hits are well read, when the gembase format is used (.prt file is
provided, prodigal is not used to find the proteins).
"""
replicon_id = 'ACBA.0917.00019'
contig_id = 'ACBA.0917.00019.0001'
result_dir_expected = self.find_data(
"Results_Integron_Finder_{}.gembase".format(replicon_id))
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', replicon_id + '.fna'))
prot_file = os.path.join(result_dir_expected,
"tmp_{}".format(contig_id),
contig_id + '.prt')
infile = os.path.join(result_dir_expected, "tmp_{}".format(contig_id),
"{}_intI.res".format(contig_id))
args = argparse.Namespace()
args.gembase = True
args.replicon = replicon_path
cfg = Config(args)
sequences_db = read_multi_prot_fasta(replicon_path)
replicon = next(sequences_db)
prot_db = GembaseDB(replicon, cfg, prot_file=prot_file)
df = read_hmm(contig_id, prot_db, infile, cfg)
exp = pd.DataFrame(data={
"Accession_number": contig_id,
"query_name": "intI_Cterm",
"ID_query": "-",
"ID_prot": "ACBA.0917.00019.i0001_00298",
"strand": -1,
"pos_beg": 311597,
"pos_end": 312631,
"evalue": 3.6e-25
},
index=[0])
exp = exp[[
"Accession_number", "query_name", "ID_query", "ID_prot", "strand",
"pos_beg", "pos_end", "evalue"
]]
pdt.assert_frame_equal(df, exp)
def test_get_description(self):
# SeqDesc(id, strand, strat, stop)
file_name = {
('ACBA.0917.00019', '.fna'): {
'ACBA.0917.00019.b0001_00001':
SeqDesc('ACBA.0917.00019.b0001_00001', -1, 266, 1480),
'ACBA.0917.00019.i0001_03957':
SeqDesc('ACBA.0917.00019.i0001_03957', -1, 4043755, 4044354)
},
}
for seq_name, ext in file_name:
replicon_path = self.find_data(
os.path.join('Gembase', 'Replicons', seq_name + ext))
self.args.replicon = replicon_path
cfg = Config(self.args)
seq_db = read_multi_prot_fasta(replicon_path)
replicon = next(seq_db)
replicon.path = replicon_path
os.makedirs(cfg.tmp_dir(replicon.id))
db = GembaseDB(replicon, cfg)
descriptions = file_name[(seq_name, ext)]
for seq_id, desc in descriptions.items():
self.assertEqual(desc, db.get_description(seq_id))
with self.assertRaises(IntegronError) as ctx:
db.get_description('nimport_naoik')
self.assertEqual(
str(ctx.exception),
"'nimport_naoik' is not a valid Gembase protein identifier.")
with self.assertRaises(KeyError) as ctx:
db.get_description('FOO.BAR.00019.i0001_03924')
self.assertEqual(str(ctx.exception), "'FOO.BAR.00019.i0001_03924'")
def test_gembase_sniffer(self):
file_names = (('ACBA.0917.00019', 'Draft'), ('ESCO001.C.00001.C001', 'Complet'))
for file_name, gem_type in file_names:
lst_path = self.find_data(os.path.join('Gembase', 'LSTINFO', file_name + '.lst'))
type_recieved = GembaseDB.gembase_sniffer(lst_path)
self.assertEqual(type_recieved, gem_type)
def find_integron_in_one_replicon(replicon, config):
"""
scan replicon for integron.
* presence of integrase
* presence of attC sites
* presence of promoters and attI sites
depending on the configuration
* perform functional annotation
produce a file containing presence of putative integrons
depending on configuration
* produce genbank file with replicon and annotations with integrons
* produce schema of replicon with integrons (in pdf)
:param replicon: the replicon to analyse.
:type replicon: a :class:`Bio.SeqRecord` object.
:param config: The configuration
:type config: a :class:`integron_finder.config.Config` object.
:returns: the path to the integron file (<replicon_id>.integrons)
and the summary file (<replicon_id.summary>).
if there is no integron the summary file is None
:rtype: tuple (str integron_file, str summary_file) or (str integron_file, None)
"""
result_tmp_dir = config.tmp_dir(replicon.id)
try:
os.mkdir(result_tmp_dir)
except OSError:
pass
tmp_replicon_path = os.path.join(result_tmp_dir, replicon.id + '.fst')
SeqIO.write(replicon, tmp_replicon_path, "fasta")
# create attr path
# used to generate protein file with prodigal
replicon.path = tmp_replicon_path
# func_annot_path is the canonical path for Functional_annotation
# path_func_annot is the path provide on the command line
if config.func_annot and not config.no_proteins and not config.path_func_annot:
if os.path.exists('bank_hmm'):
fa_hmm = scan_hmm_bank('bank_hmm')
elif os.path.exists(config.func_annot_path):
fa_hmm = scan_hmm_bank(config.func_annot_path)
else:
raise IntegronError(
"the dir '{}' neither 'bank_hmm' exists, specify the location of hmm "
"profile with --path-func-annot option".format(
config.func_annot_path))
is_func_annot = True
elif config.path_func_annot and config.no_proteins is False:
fa_hmm = scan_hmm_bank(config.path_func_annot)
is_func_annot = True
else:
is_func_annot = False
if is_func_annot and not fa_hmm:
_log.warning(
"No hmm profiles for functional annotation detected, skip functional annotation step."
)
if config.gembase_path:
protein_db = GembaseDB(replicon,
config,
gembase_path=config.gembase_path)
elif config.gembase:
protein_db = GembaseDB(replicon, config)
else:
protein_db = ProdigalDB(replicon, config)
##################
# Default search #
##################
intI_file = os.path.join(result_tmp_dir, replicon.id + "_intI.res")
phageI_file = os.path.join(result_tmp_dir, replicon.id + "_phage_int.res")
attC_default_file = os.path.join(result_tmp_dir,
replicon.id + "_attc_table.res")
try:
if not config.no_proteins:
if not os.path.isfile(intI_file) or not os.path.isfile(
phageI_file):
find_integrase(replicon.id, protein_db.protfile,
result_tmp_dir, config)
_log.info("Starting Default search ... :")
if not os.path.isfile(attC_default_file):
# find attc with cmsearch
find_attc(tmp_replicon_path,
replicon.name,
config.cmsearch,
result_tmp_dir,
config.model_attc_path,
incE=config.evalue_attc,
cpu=config.cpu)
_log.info("Default search done... : ")
integrons = find_integron(replicon, protein_db, attC_default_file,
intI_file, phageI_file, config)
#########################
# Search with local_max #
#########################
if config.local_max:
_log.info("Starting search with local_max...:")
if not os.path.isfile(
os.path.join(result_tmp_dir, "integron_max.pickle")):
circular = True if replicon.topology == 'circ' else False
integron_max = find_attc_max(
integrons,
replicon,
config.distance_threshold,
config.model_attc_path,
max_attc_size=config.max_attc_size,
min_attc_size=config.min_attc_size,
circular=circular,
out_dir=result_tmp_dir,
cpu=config.cpu,
evalue_attc=config.evalue_attc)
integron_max.to_pickle(
os.path.join(result_tmp_dir, "integron_max.pickle"))
_log.info("Search with local_max done... :")
else:
integron_max = pd.read_pickle(
os.path.join(result_tmp_dir, "integron_max.pickle"))
integron_max = integron_max[
(integron_max.evalue < config.evalue_attc)
& (abs(integron_max.pos_end -
integron_max.pos_beg) < config.max_attc_size) &
(config.min_attc_size <
abs(integron_max.pos_end - integron_max.pos_beg))]
_log.info(
"Search with local_max was already done, continue... :")
integrons = find_integron(replicon, protein_db, integron_max,
intI_file, phageI_file, config)
##########################
# Add promoters and attI #
##########################
for integron in integrons:
integron_type = integron.type()
if integron_type != "In0": # complete & CALIN
if not config.no_proteins:
_log.info("Adding proteins ... :")
integron.add_proteins(protein_db)
if config.promoter_attI:
_log.info("Adding promoters and attI ... :")
if integron_type == "complete":
integron.add_promoter()
integron.add_attI()
elif integron_type == "In0":
integron.add_attI()
integron.add_promoter()
#########################
# Functional annotation #
#########################
if is_func_annot and fa_hmm:
_log.info("Starting functional annotation ...:")
func_annot(integrons, replicon, protein_db, fa_hmm, config,
result_tmp_dir)
#######################
# Writing out results #
#######################
_log.info("Writing out results for replicon {}".format(replicon.id))
if config.pdf:
for j, integron in enumerate(integrons, 1):
if integron.type() == "complete":
integron.draw_integron(file=os.path.join(
config.result_dir, "{}_{}.pdf".format(replicon.id, j)))
base_outfile = os.path.join(config.result_dir, replicon.id)
integron_file = base_outfile + ".integrons"
_log.debug("Writing integron_file {}".format(integron_file))
if integrons:
integrons_report = results.integrons_report(integrons)
integrons_report.to_csv(integron_file,
sep="\t",
index=False,
na_rep="NA")
summary = results.summary(integrons_report)
summary_file = base_outfile + ".summary"
summary.to_csv(summary_file,
sep="\t",
na_rep="NA",
index=False,
columns=[
'ID_replicon', 'ID_integron', 'complete', 'In0',
'CALIN'
])
if config.gbk:
add_feature(replicon, integrons_report, protein_db,
config.distance_threshold)
SeqIO.write(
replicon,
os.path.join(config.result_dir, replicon.id + ".gbk"),
"genbank")
else:
with open(integron_file, "w") as out_f:
out_f.write("# No Integron found\n")
summary_file = None
except integron_finder.EmptyFileError as err:
_log.warning('############ Skip replicon {} ############'.format(
replicon.name))
integron_file = ''
summary_file = ''
#########################
# clean temporary files #
#########################
if not config.keep_tmp:
try:
shutil.rmtree(result_tmp_dir)
except Exception as err:
_log.warning("Cannot remove temporary results : '{} : {}'".format(
result_tmp_dir, str(err)))
return integron_file, summary_file