def test_generate_training_files(self):
app = RdpTaxonAssigner({
'id_to_taxonomy_fp': self.id_to_taxonomy_file.name,
'reference_sequences_fp': self.reference_seqs_file.name,
})
actual_taxonomy_file, actual_training_seqs_file = \
app._generate_training_files()
# see note in test_build_tree()
self.assertEqual(actual_taxonomy_file.read(), rdp_expected_taxonomy)
def test_parse_lineage(self):
"""Lineage in csv format is correctly parsed to a list
"""
str = 'Archaea;Euryarchaeota;Methanomicrobiales;Methanomicrobium et rel.;a;b'
actual = RdpTaxonAssigner._parse_lineage(str)
expected = ['Archaea', 'Euryarchaeota', 'Methanomicrobiales',
'Methanomicrobium et rel.', 'a', 'b']
self.assertEqual(actual, expected)
def test_build_tree(self):
"""RdpTaxonAssigner._build_tree() should return a tree with correct Rdp-format taxonomy
"""
tree = RdpTaxonAssigner._build_tree(self.id_to_taxonomy_file)
actual = tree.rdp_taxonomy()
# The order of the lines in this file depends on python's
# dict() implementation, so we should ideally build two sets
# of lines and check that their contents match.
expected = rdp_expected_taxonomy
self.assertEqual(actual, expected)
def test_parse_lineage(self):
"""Lineage in csv format is correctly parsed to a list
"""
str = 'Archaea;Euryarchaeota;Methanomicrobiales;Methanomicrobium et rel.;a;b'
actual = RdpTaxonAssigner._parse_lineage(str)
expected = [
'Archaea', 'Euryarchaeota', 'Methanomicrobiales',
'Methanomicrobium et rel.', 'a', 'b'
]
self.assertEqual(actual, expected)
def test_build_tree(self):
"""RdpTaxonAssigner._build_tree() should return a tree with correct Rdp-format taxonomy
"""
tree = RdpTaxonAssigner._build_tree(self.id_to_taxonomy_file)
actual = tree.rdp_taxonomy()
# The order of the lines in this file depends on python's
# dict() implementation, so we should ideally build two sets
# of lines and check that their contents match.
expected = rdp_expected_taxonomy
self.assertEqual(actual, expected)
def test_train_on_the_fly(self):
"""Training on-the-fly classifies reference sequence correctly with 100% certainty
"""
input_seqs_file = NamedTemporaryFile(prefix='RdpTaxonAssignerTest_',
suffix='.fasta')
input_seqs_file.write(test_seq_coll.toFasta())
input_seqs_file.seek(0)
expected = rdp_trained_test1_expected_dict
app = RdpTaxonAssigner({
'id_to_taxonomy_fp':
self.id_to_taxonomy_file.name,
'reference_sequences_fp':
self.reference_seqs_file.name,
})
actual = app(self.tmp_seq_filepath)
key = 'X67228 some description'
self.assertEqual(actual[key], expected[key])
def setUp(self):
# Temporary input file
self.tmp_seq_filepath = get_tmp_filename(
prefix='RdpTaxonAssignerTest_', suffix='.fasta')
seq_file = open(self.tmp_seq_filepath, 'w')
seq_file.write(rdp_test1_fasta)
seq_file.close()
# Temporary results filename
self.tmp_res_filepath = get_tmp_filename(
prefix='RdpTaxonAssignerTestResult_',
suffix='.tsv',
)
# touch the file so we don't get an error trying to close it
open(self.tmp_res_filepath, 'w').close()
# Temporary log filename
self.tmp_log_filepath = get_tmp_filename(
prefix='RdpTaxonAssignerTestLog_',
suffix='.txt',
)
# touch the file so we don't get an error trying to close it
open(self.tmp_log_filepath, 'w').close()
self._paths_to_clean_up = \
[self.tmp_seq_filepath, self.tmp_res_filepath, self.tmp_log_filepath]
self.id_to_taxonomy_file = NamedTemporaryFile(
prefix='RdpTaxonAssignerTest_', suffix='.txt')
self.id_to_taxonomy_file.write(rdp_id_to_taxonomy)
self.id_to_taxonomy_file.seek(0)
self.reference_seqs_file = NamedTemporaryFile(
prefix='RdpTaxonAssignerTest_', suffix='.fasta')
self.reference_seqs_file.write(rdp_reference_seqs)
self.reference_seqs_file.seek(0)
self.default_app = RdpTaxonAssigner({})
def test_generate_training_seqs(self):
seqs = RdpTaxonAssigner._generate_training_seqs(
self.reference_seqs_file, self.id_to_taxonomy_file)
actual = LoadSeqs(data=seqs, aligned=False).toFasta()
self.assertEqual(actual, rdp_expected_training_seqs)
def test_init(self):
"""RdpTaxonAssigner.__init__ should set default attributes and params
"""
a = RdpTaxonAssigner({})
self.assertEqual(a.Name, 'RdpTaxonAssigner')
def test_generate_training_seqs(self):
seqs = RdpTaxonAssigner._generate_training_seqs(
self.reference_seqs_file, self.id_to_taxonomy_file)
actual = LoadSeqs(data=seqs, aligned=False).toFasta()
self.assertEqual(actual, rdp_expected_training_seqs)
