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 convert_ps_to_pdf(config, executor):
after = combine_ps_files(config, executor)
combined_ps_name = config['combined_ps_name']
pdf_filename = replace_ext(combined_ps_name, 'pdf')
config['pdf_filename'] = pdf_filename
config['allplots_result'] = pdf_filename
return enqueue(_ps2pdf, executor, config, pdf_filename, after=after)
def plan(config, executor):
if parsing.isgbk(config):
logger.debug(
"GBK file, extracting known gene names %s", config['filename'])
rconfig, hash = get_hash(config)
target_file = parsing.derive_filename(config, hash, 'genes')
config[OUTPUTKEY] = target_file
return enqueue(_extract, executor, rconfig, target_file)
def plan(config, executor):
if parsing.isgbk(config):
logger.debug(
"GBK file, extracting known CDS to json %s", config['filename'])
rconfig, hash = get_hash(config)
target_file = parsing.derive_filename(config, hash, 'track.genes.json')
config['InputCDSFileJson'] = target_file
return enqueue(_extract, executor, rconfig, target_file)
def combine_ps_files(config, executor):
after = allplots(config, executor)
psnames = config['psnames']
log.debug("Going to combine %d postscript files", len(psnames))
combined_ps_name = parsing.derive_filename(
config, Hasher().hashlist(psnames).hexdigest(), 'ps')
config['combined_ps_name'] = combined_ps_name
return enqueue(
_combine_ps_files, executor, config, combined_ps_name, after=after)
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
def allplots(config, executor):
after = []
try:
after.extend(extract.plan(config, executor))
except:
pass
after.extend(nprofile.plan(config, executor))
after.extend(acgt_gamma.plan(config, executor))
parsing.length(config)
parsing.first_page_title(config)
parsing.following_page_title(config)
parsing.endBase(config)
h = Hasher()
# Strip down to the config for this task only
rconfig = reducedict(config, KEYS + FILE_KEYS)
basesPerGraph = rconfig['basesPerGraph']
graphsPerPage = rconfig['graphsPerPage']
startBase = rconfig.pop('startBase')
endBase = rconfig.pop('endBase')
bp_per_page = rconfig['bp_per_page'] = basesPerGraph * graphsPerPage
page_count = math.ceil(float(endBase - startBase) / bp_per_page)
log.info("Generating %d pages of allplots", page_count)
page_num = 1 # page number offset
filenames = []
waiton = []
# per-page loop
while startBase < endBase:
pconfig = dict(rconfig.items())
pconfig['page_num'] = page_num
pconfig['startBase'] = startBase
if startBase + bp_per_page < endBase:
pconfig['endBase'] = startBase + bp_per_page
else:
pconfig['endBase'] = endBase
h = Hasher().hashdict(pconfig).hashfiletime(BIN).hashfiletime(__file__)
psname = parsing.derive_filename(config, h.hexdigest(), 'ps')
filenames.append(psname)
waiton.extend(enqueue(_ap, executor, pconfig, psname, after=after))
page_num += 1
startBase += bp_per_page
# Finally set the output filenames into the master config dict
config['psnames'] = filenames
return waiton