def plan(config, executor):
"Identifying ORFs with significant 3-base periodicities."
if config.get('skip_prediction', False):
return
assert os.path.exists(DATAPATH), \
"Missing pynpact/data for acgt_gamma prediction. " \
"Expected at " + DATAPATH
rconfig = reducedict(config, [
'filename', 'significance', 'GeneDescriptorSkip1', 'mycoplasma', 'stderr'])
h = Hasher().hashdict(rconfig)
h.hashfiletime(config['filename'])
h.hashfiletime(BIN)
outdir = parsing.derive_filename(config, h.hexdigest(), '.predict')
log.debug("Adding prediction filenames to config dict.")
# strip 4 characters off here b/c that's how acgt_gamma does
# it at about lines 262-270
j = lambda ext: os.path.join(
outdir, os.path.basename(config['filename'])[:-4] + ext)
config['NewOrfsFile'] = config['File_of_new_CDSs'] = j(".newcds")
config['ModifiedOrfsFile'] = config['File_of_published_rejected_CDSs'] = j(".modified")
config['HitsFile'] = config['File_of_G+C_coding_potential_regions'] = j('.profiles')
config[OUTPUTKEY] = outdir
return enqueue(_acgt_gamma, executor, rconfig, outdir)
def get_hash(config):
config = reducedict(config, KEYS)
h = Hasher()
h.hashdict(config)
h.hashfiletime(BIN)
h.hashfiletime(config['filename'])
return config, h.hexdigest()
def plan(config, executor):
if 'nprofile' in config:
return
config['nprofile'] = True
parsing.length(config)
rconfig = reducedict(config, KEYS)
h = Hasher()
h.hashdict(rconfig)
h.hashfiletime(BIN)
hash = h.hexdigest()
target = parsing.derive_filename(config, hash, 'nprofile')
config[OUTPUTKEY] = target
config[JSONOUTPUTKEY] = target + '.json'
jobs = enqueue(_nprofile, executor, rconfig, target)
enqueue(_nprofile_to_json, executor, {OUTPUTKEY: target},
config[JSONOUTPUTKEY], after=jobs)
return jobs
