def test_tabular_output(self):
""" SortMeRNA should output a BLAST tabular output
"""
sortmerna_db, db_files_to_remove = build_database_sortmerna(
abspath(self.file_reference_seq_fp),
max_pos=250,
output_dir=self.output_dir)
self.files_to_remove.extend(db_files_to_remove)
# Run SortMeRNA
clusters, failures, smr_files_to_remove = sortmerna_ref_cluster(
seq_path=self.file_read_seqs_fp,
sortmerna_db=sortmerna_db,
refseqs_fp=self.file_reference_seq_fp,
result_path=join(self.output_dir, "sortmerna_otus.txt"),
tabular=True)
self.assertTrue(exists(join(self.output_dir,
"sortmerna_otus.blast")))
def test_sortmerna_default_param(self):
""" SortMeRNA version 2.0 reference OTU picking works with default settings
"""
# rebuild the index
sortmerna_db, db_files_to_remove = build_database_sortmerna(
abspath(self.file_reference_seq_fp),
max_pos=250,
output_dir=self.output_dir)
# Files created by indexdb_rna to be deleted
self.files_to_remove.extend(db_files_to_remove)
# Run SortMeRNA
cluster_map, failures, smr_files_to_remove = sortmerna_ref_cluster(
seq_path=self.file_read_seqs_fp,
sortmerna_db=sortmerna_db,
refseqs_fp=self.file_reference_seq_fp,
result_path=join(self.output_dir, "sortmerna_otus.txt"))
# Check all sortmerna output files exist
output_files = [join(self.output_dir, ext)
for ext in ['sortmerna_otus_otus.txt',
'sortmerna_otus.log',
'sortmerna_otus_denovo.fasta',
'sortmerna_otus.fasta']]
# Check output files exist
for fp in output_files:
self.assertTrue(exists(fp))
# Files created sortmerna to be deleted (StdErr and StdOut were already
# removed in sortmerna_ref_cluster)
self.files_to_remove.extend(output_files)
# Random reads that should not appear in any output file
random_reads = ['simulated_random_reads.fa.000000000',
'simulated_random_reads.fa.000000001',
'simulated_random_reads.fa.000000002',
'simulated_random_reads.fa.000000003',
'simulated_random_reads.fa.000000004',
'simulated_random_reads.fa.000000005',
'simulated_random_reads.fa.000000006',
'simulated_random_reads.fa.000000007',
'simulated_random_reads.fa.000000008',
'simulated_random_reads.fa.000000009']
# Reads passing E-value threshold and with similarity/coverage >=97%
otu_reads = ['HMPMockV1.2.Staggered2.673827_47',
'HMPMockV1.2.Staggered2.673827_115',
'HMPMockV1.2.Staggered2.673827_122',
'HMPMockV1.2.Staggered2.673827_161',
'HMPMockV1.2.Staggered2.673827_180',
'HMPMockV1.2.Staggered2.673827_203',
'HMPMockV1.2.Staggered2.673827_207',
'HMPMockV1.2.Staggered2.673827_215',
'HMPMockV1.2.Staggered2.673827_218',
'HMPMockV1.2.Staggered2.673827_220']
# Reads passing E-value threshold and with similarity/coverage <97%
denovo_reads = ['HMPMockV1.2.Staggered2.673827_0',
'HMPMockV1.2.Staggered2.673827_1',
'HMPMockV1.2.Staggered2.673827_2',
'HMPMockV1.2.Staggered2.673827_3',
'HMPMockV1.2.Staggered2.673827_4',
'HMPMockV1.2.Staggered2.673827_5',
'HMPMockV1.2.Staggered2.673827_6',
'HMPMockV1.2.Staggered2.673827_7',
'HMPMockV1.2.Staggered2.673827_8',
'HMPMockV1.2.Staggered2.673827_9']
# Check correct number of OTU clusters in file
otu_clusters = ['295053']
f_aligned = open(output_files[3], "U")
f_otumap = open(output_files[0], "U")
f_denovo = open(output_files[2], "U")
# Verify the aligned FASTA file
for label, seq in parse_fasta(f_aligned):
id = label.split()[0]
# Read is not random
self.assertNotIn(id, random_reads)
# Read is either in otu_reads or denovo_reads
self.assertIn(id, otu_reads+denovo_reads)
f_aligned.close()
# Verify the de novo reads FASTA file
for label, seq in parse_fasta(f_denovo):
id = label.split()[0]
# Read is not random
self.assertNotIn(id, random_reads)
# Read is not an OTU read
self.assertNotIn(id, otu_reads)
# Read is a de novo read
self.assertIn(id, denovo_reads)
f_denovo.close()
# Check the OTU map
for line in f_otumap:
otu_entry = line.split()
# Cluster ID is correct
self.assertIn(otu_entry[0], otu_clusters)
# Each read in the cluster must exclusively be an OTU read
for read in otu_entry[1:]:
self.assertNotIn(read, random_reads)
self.assertNotIn(read, denovo_reads)
self.assertIn(read, otu_reads)
f_otumap.close()
# Check returned list of lists of clusters
expected_cluster = ['HMPMockV1.2.Staggered2.673827_47',
'HMPMockV1.2.Staggered2.673827_115',
'HMPMockV1.2.Staggered2.673827_122',
'HMPMockV1.2.Staggered2.673827_161',
'HMPMockV1.2.Staggered2.673827_180',
'HMPMockV1.2.Staggered2.673827_203',
'HMPMockV1.2.Staggered2.673827_207',
'HMPMockV1.2.Staggered2.673827_215',
'HMPMockV1.2.Staggered2.673827_218',
'HMPMockV1.2.Staggered2.673827_220']
# Should only have 1 cluster
self.assertEqual(1, len(cluster_map))
for actual_cluster in cluster_map.itervalues():
actual_cluster.sort()
expected_cluster.sort()
self.assertEqual(actual_cluster, expected_cluster)
# Check log file number of clusters and failures corresponds to
# the results in the output files
f_log = open(output_files[1], "U")
num_clusters = 0
num_failures = 0
for line in f_log:
if line.startswith(" Total OTUs"):
num_clusters = (re.split(' = ', line)[1]).strip()
elif line.startswith(" Total reads for de novo clustering"):
num_failures = (re.split(' = ', line)[1]).strip()
f_log.close()
self.assertEqual(int(num_clusters), len(otu_clusters))
self.assertEqual(int(num_failures), len(denovo_reads))
