def main(argv=None):
opts = parse_args(argv)
# Configure
if opts.verbose:
logging.basicConfig(level=logging.DEBUG)
else:
logging.basicConfig(level=logging.WARNING)
taxa_db = NcbiEutils(opts.cache_fp)
taxa_db.load_cache()
consensus_thresholds = [t for _, t in CONSENSUS_THRESHOLDS]
assigner = Assigner(
opts.min_cover, opts.min_species_id, opts.min_genus_id, opts.min_id,
consensus_thresholds, opts.max_generic, taxa_db)
# Read input files
with open(opts.fasta_file) as f:
sequences = list(iter_fasta(f))
with open(opts.blast_file) as f:
blast_hits = read_blast(f)
# Open output files
if not os.path.exists(opts.output_directory):
os.mkdir(opts.output_directory)
output_file = open(
os.path.join(opts.output_directory, "Full_Taxonomy.txt"), 'w')
standard_taxa_file = open(
os.path.join(opts.output_directory, "Standard_Taxonomy.txt"), "w")
log_file = open(os.path.join(opts.output_directory, "brocc.log"), "w")
log_file.write(
"Sequence\tWinner_Votes\tVotes_Cast\tGenerics_Pruned\tLevel\t"
"Classification\n")
# Do the work
for name, seq in sequences:
seq_hits = blast_hits[name]
# This is where the magic happens
a = assigner.assign(name, seq, seq_hits)
output_file.write(a.format_for_full_taxonomy())
standard_taxa_file.write(a.format_for_standard_taxonomy())
log_file.write(a.format_for_log())
# Close output files, write cache
output_file.close()
standard_taxa_file.close()
log_file.close()
taxa_db.save_cache()
def test_save_load_cache(self):
lineages = {
"taxon1": {'class': "a", "genus": "b"},
"taxon2": {'class': "c", "genus": "d"},
}
taxon_ids = {"taxon1": "b", "taxon2": "d"}
self.db.lineages = lineages
self.db.taxon_ids = taxon_ids
self.db._fresh = False
self.db.save_cache()
db2 = NcbiEutils(self.cache_file.name)
db2.load_cache()
self.assertEqual(db2.lineages, lineages)
self.assertEqual(db2.taxon_ids, taxon_ids)
class NcbiEutilsTests(unittest.TestCase):
def setUp(self):
self.cache_file = tempfile.NamedTemporaryFile(suffix=".json")
self.db = NcbiEutils(self.cache_file.name)
def test_save_load_cache(self):
lineages = {
"taxon1": {'class': "a", "genus": "b"},
"taxon2": {'class': "c", "genus": "d"},
}
taxon_ids = {"taxon1": "b", "taxon2": "d"}
self.db.lineages = lineages
self.db.taxon_ids = taxon_ids
self.db._fresh = False
self.db.save_cache()
db2 = NcbiEutils(self.cache_file.name)
db2.load_cache()
self.assertEqual(db2.lineages, lineages)
self.assertEqual(db2.taxon_ids, taxon_ids)
def test_get_taxon_id(self):
self.assertEqual(self.db.get_taxon_id("312434489"), "531911")
self.assertEqual(self.db.taxon_ids, {"312434489": "531911"})
def test_get_lineage(self):
observed_lineage = self.db.get_lineage("531911")
expected_lineage = {
'Lineage': (
'cellular organisms; Eukaryota; Opisthokonta; Fungi; Dikarya; '
'Ascomycota; saccharomyceta; Pezizomycotina; leotiomyceta; '
'sordariomyceta; Sordariomycetes; Xylariomycetidae; '
'Xylariales; Amphisphaeriaceae; Pestalotiopsis'),
'class': 'Sordariomycetes',
'family': 'Amphisphaeriaceae',
'genus': 'Pestalotiopsis',
'kingdom': 'Fungi',
'no rank': 'sordariomyceta',
'order': 'Xylariales',
'phylum': 'Ascomycota',
'species': 'Pestalotiopsis maculiformans',
'subclass': 'Xylariomycetidae',
'subkingdom': 'Dikarya',
'subphylum': 'Pezizomycotina',
'superkingdom': 'Eukaryota',
}
self.assertEqual(observed_lineage, expected_lineage)
self.assertEqual(self.db.lineages, {'531911': expected_lineage})
def test_get_lineage(self):
db = NcbiEutils()
observed_lineage = db.get_lineage("531911")
expected_lineage = [('cellular organisms', 'no rank'),
('Eukaryota', 'superkingdom'),
('Opisthokonta', 'no rank'), ('Fungi', 'kingdom'),
('Dikarya', 'subkingdom'),
('Ascomycota', 'phylum'),
('saccharomyceta', 'no rank'),
('Pezizomycotina', 'subphylum'),
('leotiomyceta', 'no rank'),
('sordariomyceta', 'no rank'),
('Sordariomycetes', 'class'),
('Xylariomycetidae', 'subclass'),
('Xylariales', 'order'),
('Sporocadaceae', 'family'),
('Pestalotiopsis', 'genus'),
('Pestalotiopsis maculiformans', 'species')]
self.assertEqual(observed_lineage, expected_lineage)
self.assertEqual(db.lineages, {'531911': expected_lineage})
def main(argv=None):
opts = parse_args(argv)
# Configure
if opts.verbose:
logging.basicConfig(level=logging.DEBUG)
else:
logging.basicConfig(level=logging.WARNING)
taxa_db = NcbiEutils(opts.cache_fp)
taxa_db.load_cache()
consensus_thresholds = [t for _, t in CONSENSUS_THRESHOLDS]
assigner = Assigner(opts.min_cover, opts.min_species_id, opts.min_genus_id,
opts.min_id, consensus_thresholds, opts.max_generic,
taxa_db)
# Read input files
with open(opts.fasta_file) as f:
sequences = list(iter_fasta(f))
with open(opts.blast_file) as f:
blast_hits = read_blast(f)
# Open output files
if not os.path.exists(opts.output_directory):
os.mkdir(opts.output_directory)
output_file = open(
os.path.join(opts.output_directory, "Full_Taxonomy.txt"), 'w')
standard_taxa_file = open(
os.path.join(opts.output_directory, "Standard_Taxonomy.txt"), "w")
log_file = open(os.path.join(opts.output_directory, "brocc.log"), "w")
log_file.write(
"Sequence\tWinner_Votes\tVotes_Cast\tGenerics_Pruned\tLevel\t"
"Classification\n")
# Do the work
for name, seq in sequences:
seq_hits = blast_hits[name]
# This is where the magic happens
a = assigner.assign(name, seq, seq_hits)
output_file.write(a.format_for_full_taxonomy())
standard_taxa_file.write(a.format_for_standard_taxonomy())
log_file.write(a.format_for_log())
# Close output files, write cache
output_file.close()
standard_taxa_file.close()
log_file.close()
taxa_db.save_cache()
def test_get_taxon_id(self):
db = NcbiEutils()
self.assertEqual(db.get_taxon_id("HQ608011.1"), "531911")
self.assertEqual(db.taxon_ids, {"HQ608011.1": "531911"})
def setUp(self):
self.cache_file = tempfile.NamedTemporaryFile(suffix=".json")
self.db = NcbiEutils(self.cache_file.name)
def main(argv=None):
opts = parse_args(argv)
# Configure
if opts.verbose:
logging.basicConfig(level=logging.DEBUG)
else:
logging.basicConfig(level=logging.WARNING)
if os.path.exists(opts.taxonomy_db):
taxa_db = NcbiLocal(opts.taxonomy_db)
else:
sys.stderr.write(
"Did not detect a local copy of the NCBI taxonomy.\n"
"Using NCBI EUtils to get taxonomic info instead.\n\n"
"The NCBI taxonomy can be dowloaded with the script "
"create_local_taxonomy_db.py\n"
"This will greatly speed up the assignment process.\n")
taxa_db = NcbiEutils()
consensus_thresholds = [t for _, t in CONSENSUS_THRESHOLDS]
assigner = Assigner(opts.min_cover, opts.min_species_id, opts.min_genus_id,
opts.min_id, consensus_thresholds,
opts.min_winning_votes, taxa_db)
# Read input files
with open(opts.fasta_file) as f:
sequences = list(iter_fasta(f))
with open(opts.blast_file) as f:
blast_hits = read_blast(f)
# Open output files
if not os.path.exists(opts.output_directory):
os.mkdir(opts.output_directory)
standard_taxa_file = open(
os.path.join(opts.output_directory, "Standard_Taxonomy.txt"), "w")
log_file = open(os.path.join(opts.output_directory, "brocc.log"), "w")
log_file.write(
"Sequence\tWinner_Votes\tVotes_Cast\tGenerics_Pruned\tLevel\t"
"Classification\n")
# Set up log for voting details
vote_logger = logging.getLogger("brocc.votes")
vote_logger.setLevel(logging.DEBUG)
vote_handler = logging.FileHandler(
os.path.join(opts.output_directory, "voting_log.txt"))
vote_handler.setLevel(logging.DEBUG)
vote_formatter = logging.Formatter('%(message)s')
vote_handler.setFormatter(vote_formatter)
vote_logger.addHandler(vote_handler)
vote_logger.propagate = False
# Do the work
for name, seq in sequences:
seq_hits = blast_hits[name]
# This is where the magic happens
a = assigner.assign(name, seq, seq_hits)
standard_taxa_file.write(a.format_for_standard_taxonomy())
log_file.write(a.format_for_log())
# Close output files
standard_taxa_file.close()
log_file.close()