def test_parallel_rdp_taxonomy_assigner(self):
""" parallel_rdp_taxonomy_assigner functions as expected """
params = {
'id_to_taxonomy_fp': self.id_to_taxonomy_file.name,
'rdp_max_memory': 1500,
'rdp_classifier_fp': getenv('RDP_JAR_PATH'),
'confidence': 0.80,
'reference_seqs_fp': self.reference_seqs_file.name
}
app = ParallelRdpTaxonomyAssigner()
r = app(self.tmp_seq_filepath,
self.test_out,
params,
job_prefix='RDPTEST',
poll_directly=True,
suppress_submit_jobs=False)
results = fields_to_dict(
open(glob(join(self.test_out, '*_tax_assignments.txt'))[0], 'U'))
# some basic sanity checks: we should get the same number of sequences
# as our input with the same seq IDs. We should have a taxonomy string
# and a confidence value for each seq as well.
self.assertEqual(len(results), 2)
self.assertEqual(len(results['X67228']), 2)
self.assertEqual(len(results['EF503697']), 2)
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
# create dict of command-line options
params = eval(str(opts))
parallel_runner = ParallelRdpTaxonomyAssigner(
cluster_jobs_fp=opts.cluster_jobs_fp,
jobs_to_start=opts.jobs_to_start,
retain_temp_files=opts.retain_temp_files,
suppress_polling=opts.suppress_polling,
seconds_to_sleep=opts.seconds_to_sleep)
parallel_runner(opts.input_fasta_fp,
opts.output_dir,
params,
job_prefix=opts.job_prefix,
poll_directly=opts.poll_directly,
suppress_submit_jobs=opts.suppress_submit_jobs)