def test_get_ids(self):
'''test _get_ids'''
infile = os.path.join(data_dir, 'cdhit_test_get_ids.fa')
expected = {'id1', 'id2', 'id3'}
r = cdhit.Runner(infile, 'out')
got = r._get_ids(infile)
self.assertEqual(expected, got)
def test_get_run_cmd_with_unlimited_memory(self):
'''test_get_run_cmd_with_unlimited_memory'''
fa_infile = os.path.join(data_dir, 'cdhit_test_run_get_clusters_from_dict_rename.in.fa')
r = cdhit.Runner(fa_infile, memory_limit=0)
run_cmd = r.get_run_cmd('foo/bar/file.out')
match = re.search('^.+ -o foo/bar/file.out -c 0.9 -T 1 -s 0.0 -d 0 -bak 1 -M 0$', run_cmd)
self.assertIsNotNone(match, msg="Command output was " + run_cmd)
def test_fake_run_fail(self):
'''test fake_run with non-unique names'''
infile = os.path.join(data_dir, 'cdhit_test_fake_run.non-unique.in.fa')
tmpfile = 'tmp.cdhit_test_fake_run.out.non-unique.fa'
r = cdhit.Runner(infile, tmpfile)
with self.assertRaises(cdhit.Error):
clusters = r.fake_run()
os.unlink(tmpfile)
def test_run_min_cluster_number_42(self):
'''test run with min_cluster_number 42'''
infile = os.path.join(data_dir, 'cdhit_test_run.in.fa')
r = cdhit.Runner(infile, min_cluster_number=42)
clusters = r.run()
expected_clusters = {
'42': {'seq1', 'seq2', 'seq3'},
'43': {'seq4'},
}
self.assertEqual(clusters, expected_clusters)
def test_run(self):
'''test run'''
infile = os.path.join(data_dir, 'cdhit_test_run.in.fa')
r = cdhit.Runner(infile)
clusters = r.run()
expected_clusters = {
'0': {'seq1', 'seq2', 'seq3'},
'1': {'seq4'},
}
self.assertEqual(clusters, expected_clusters)
def test_rename_fasta(self):
'''test _rename_fasta'''
infile = os.path.join(data_dir, 'cdhit_test_rename_fasta.in.fa')
tmpfile = 'tmp.rename_fasta.out.fa'
expected = os.path.join(data_dir, 'cdhit_test_rename_fasta.out.fa')
names_dict = {'a': 'seq1', 'b': 'seq2', 'c': 'seq3'}
r = cdhit.Runner(infile, 'out')
r._rename_fasta(infile, tmpfile, names_dict)
self.assertTrue(filecmp.cmp(expected, tmpfile, shallow=False))
os.unlink(tmpfile)
def test_run_get_clusters_from_file(self):
'''test run_get_clusters_from_file'''
fa_infile = os.path.join(data_dir, 'cdhit_test_run_get_clusters_from_dict.in.fa')
clusters_infile = os.path.join(data_dir, 'cdhit_test_run_get_clusters_from_dict.in.clusters')
r = cdhit.Runner(fa_infile)
clusters = r.run_get_clusters_from_file(clusters_infile, {'seq1', 'seq2', 'seq3'})
expected_clusters = {
'0': {'seq1', 'seq2'},
'1': {'seq3'},
}
self.assertEqual(clusters, expected_clusters)
def test_run_get_clusters_from_file_with_renaming(self):
'''test run_get_clusters_from_file with renaming'''
rename_dict = {'seq2': 'seq2_renamed'}
fa_infile = os.path.join(data_dir, 'cdhit_test_run_get_clusters_from_dict_rename.in.fa')
clusters_infile = os.path.join(data_dir, 'cdhit_test_run_get_clusters_from_dict.in.clusters')
r = cdhit.Runner(fa_infile)
clusters = r.run_get_clusters_from_file(clusters_infile, {'seq1', 'seq2_renamed', 'seq3'}, rename_dict=rename_dict)
expected_clusters = {
'0': {'seq1', 'seq2_renamed'},
'1': {'seq3'},
}
self.assertEqual(clusters, expected_clusters)
def test_enumerate_fasta(self):
'''test _enumerate_fasta'''
infile = os.path.join(data_dir, 'cdhit_test_enumerate_fasta.in.fa')
expected_outfile = os.path.join(data_dir,
'cdhit_test_enumerate_fasta.out.fa')
tmpfile = 'tmp.test_enumerate_fasta.out.fa'
expected_dict = {'1': 'a', '2': 'b', '3': 'c'}
r = cdhit.Runner(infile, 'out')
got_dict = r._enumerate_fasta(infile, tmpfile)
self.assertTrue(filecmp.cmp(expected_outfile, tmpfile, shallow=False))
self.assertEqual(expected_dict, got_dict)
os.unlink(tmpfile)
def test_run(self):
'''test run'''
infile = os.path.join(data_dir, 'cdhit_test_run.in.fa')
expected_outfile = os.path.join(data_dir, 'cdhit_test_run.out.fa')
tmpfile = 'tmp.cdhit_test_run.out.fa'
r = cdhit.Runner(infile, tmpfile)
clusters = r.run()
expected_clusters = {
'0': {'seq1', 'seq2', 'seq3'},
'1': {'seq4'},
}
self.assertEqual(clusters, expected_clusters)
self.assertTrue(filecmp.cmp(tmpfile, expected_outfile, shallow=False))
os.unlink(tmpfile)
def cluster_with_cdhit(self,
outprefix,
seq_identity_threshold=0.9,
threads=1,
length_diff_cutoff=0.0,
memory_limit=None,
nocluster=False,
verbose=False,
clusters_file=None):
clusters = {}
ReferenceData._write_sequences_to_files(self.sequences, self.metadata,
outprefix)
ref_types = ('noncoding', 'noncoding.varonly', 'gene', 'gene.varonly')
for ref_type in ref_types:
ref_file = outprefix + '.' + ref_type + '.fa'
if os.path.getsize(ref_file) == 0:
continue
if len(clusters) == 0:
min_cluster_number = 0
else:
min_cluster_number = 1 + max([int(x) for x in clusters.keys()])
cdhit_runner = cdhit.Runner(
ref_file,
seq_identity_threshold=seq_identity_threshold,
threads=threads,
length_diff_cutoff=length_diff_cutoff,
memory_limit=memory_limit,
verbose=verbose,
min_cluster_number=min_cluster_number,
)
if clusters_file is not None:
new_clusters = cdhit_runner.run_get_clusters_from_file(
clusters_file,
self.sequences,
rename_dict=self.rename_dict)
elif nocluster:
new_clusters = cdhit_runner.fake_run()
else:
new_clusters = cdhit_runner.run()
clusters.update(new_clusters)
self.write_cluster_allocation_file(clusters,
outprefix + '.clusters.tsv')
return clusters
def test_parse_cluster_info_file(self):
'''test _parse_cluster_info_file'''
infile = os.path.join(data_dir,
'cdhit_test_parse_cluster_info_file.in.fa')
r = cdhit.Runner(infile, 'out')
names_dict = {str(i): 'seq' + str(i) for i in range(1, 5)}
cluster_representatives = {'1', '4'}
cluster_file = os.path.join(
data_dir, 'cdhit_test_parse_cluster_info_file.out.fa.bak.clstr')
got_clusters, got_reps = r._parse_cluster_info_file(
cluster_file, names_dict, cluster_representatives)
expected_clusters = {'0': {'seq1', 'seq2', 'seq3'}, '1': {'seq4'}}
expected_reps = {'1': '0', '4': '1'}
self.assertEqual(expected_clusters, got_clusters)
self.assertEqual(expected_reps, got_reps)
def _run_cdhit(self):
r = cdhit.Runner(
self.db_fasta,
self.db_fasta_clustered,
seq_identity_threshold=self.cdhit_seq_identity_threshold,
threads=self.threads,
length_diff_cutoff=self.cdhit_length_diff_cutoff,
verbose=self.verbose,
)
if self.run_cd_hit:
self.cluster_ids = r.run()
else:
if self.verbose:
print('Skipping cd-hit because --no_cdhit option used')
self.cluster_ids = r.fake_run()
def test_fake_run_fail(self):
'''test fake_run with non-unique names'''
infile = os.path.join(data_dir, 'cdhit_test_fake_run.non-unique.in.fa')
r = cdhit.Runner(infile)
with self.assertRaises(cdhit.Error):
r.fake_run()
def test_init_fail_invalid_memory(self):
'''test_init_fail_invalid_memory'''
infile = os.path.join(data_dir, 'cdhit_test_run.in.fa')
with self.assertRaises(cdhit.Error):
cdhit.Runner(infile, memory_limit=-10)
def test_init_fail_infile_missing(self):
'''test init_fail_infile_missing'''
with self.assertRaises(cdhit.Error):
cdhit.Runner('oopsnotafile', 'out')
