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 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_missing_read(self):
obs = read_blast(StringIO(normal_output))
self.assertEqual(obs['sdlkj'], [])
def test_malformed_output(self):
obs = read_blast(StringIO(malformed_output))
h = obs['0 E7_168192'][0]
self.assertEqual(h.accession, "GQ513762.1")
self.assertEqual(h.pct_id, 98.74)
self.assertEqual(h.length, 159)
def test_malformed_output(self):
obs = read_blast(StringIO(malformed_output))
h = obs['0 E7_168192'][0]
self.assertEqual(h.gi, "259100874")
self.assertEqual(h.pct_id, 98.74)
self.assertEqual(h.length, 159)
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()
def test_missing_read(self):
obs = read_blast(StringIO(normal_output))
self.assertEqual(obs['sdlkj'], [])
def test_malformed_output(self):
obs = read_blast(StringIO(malformed_output))
h = obs['0 E7_168192'][0]
self.assertEqual(h.gi, "259100874")
self.assertEqual(h.pct_id, 98.74)
self.assertEqual(h.length, 159)