def test_parallel_database_mapper_usearch(self):
""" parallel_database_mapper_usearch functions as expected """
params = {
'refseqs_fp': self.refseqs1_fp,
'min_percent_id': 0.97,
'evalue': 1e-10,
'max_accepts': 1,
'max_rejects': 32,
'queryalnfract': 0.35,
'targetalnfract': 0.0,
'observation_metadata_fp': None
}
app = ParallelDatabaseMapperUsearch()
r = app(self.inseqs1_fp,
self.test_out,
params,
job_prefix='PTEST',
poll_directly=True,
suppress_submit_jobs=False)
observation_map_fp = glob(join(self.test_out,
'observation_map.txt'))[0]
omap = parse_otu_map(open(observation_map_fp, 'U'))
self.assertEqual(len(omap[0]), 3)
self.assertItemsEqual(
omap[1], ['eco:b0015', 'eco:b0122', 'eco:b0015:duplicate'])
self.assertItemsEqual(omap[2], ['eco:b0015-pr', 'eco:b0122-pr'])
def main():
option_parser, opts, args =\
parse_command_line_parameters(**script_info)
params = eval(str(opts))
params['save_uc_files'] = True
if opts.assignment_method == 'usearch':
parallel_runner = ParallelDatabaseMapperUsearch(
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_seqs_filepath,
opts.output_dir,
params,
job_prefix=opts.job_prefix,
poll_directly=opts.poll_directly,
suppress_submit_jobs=opts.suppress_submit_jobs)
elif opts.assignment_method == 'blat':
parallel_runner = ParallelDatabaseMapperBlat(
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_seqs_filepath,
opts.output_dir,
params,
job_prefix=opts.job_prefix,
poll_directly=opts.poll_directly,
suppress_submit_jobs=opts.suppress_submit_jobs)
elif opts.assignment_method == 'bwa-short':
# cast max_diff to an int if it's being passed as an int
if params['max_diff'] is not None and params['max_diff'] > 1.0:
params['max_diff'] = int(params['max_diff'])
parallel_runner = ParallelDatabaseMapperBwaShort(
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_seqs_filepath,
opts.output_dir,
params,
job_prefix=opts.job_prefix,
poll_directly=opts.poll_directly,
suppress_submit_jobs=opts.suppress_submit_jobs)
else:
# other -- shouldn't be able to get here as a KeyError would have
# been raised earlier
raise ValueError(
"Unknown read mapping method: %s" %
opts.assignment_method)