short_name = ALIASES[genome]
outdirect = os.path.join(RESULTSDIR, 'sampling', short_name)
if not os.path.exists(outdirect):
os.mkdir(outdirect)
ifile = os.path.join(DATADIR, genome + '.fa')
for i in xrange(options.numsamples):
outfile = os.path.join(outdirect, short_name + '%05d.txt' % (i+1,))
if os.path.exists(outfile) and not options.forcenew:
continue
logging.warning('Processing subsample %d' % (i+1,))
fgen = SeqGen(ifile)
sub_gen = SampleIterator(fgen, options.percentage)
d = DictFromGen(sub_gen, label = short_name, chunk_size = 10)
logging .warning('Writting file: %s' % outfile)
with open(outfile, 'w') as handle:
l = d.keys()
l.sort()
for key in l:
count, frac = d[key]
elm, spec = key
handle.write('\t'.join(map(str, [elm, spec, count, '%.10f' % frac]))+'\n')
if __name__ == '__main__':
parser = OptionParser()
parser.add_option('-c', '--use-cloud', default = False, action = 'store_true',
dest = 'usecloud', help = 'Use PiCloud computing')
(options, args) = parser.parse_args()
cloud.setkey(CLOUD_KEY, CLOUD_SECRET)
if not options.usecloud: cloud.start_simulator()
for short_name in args:
logging.warning('Processing Genome: %s' % short_name)
flu_gen = GetFluSeqs(organism = FLU_NAMES[short_name])
outdata = DictFromGen(flu_gen, label = short_name)
logging.warning('writting data for %s' % short_name)
with open(os.path.join(RESULTSDIR, 'flu_elmdict_'+short_name), 'w') as handle:
l = outdata.keys()
l.sort()
for key in l:
count, frac = outdata[key]
elm, spec = key
handle.write('\t'.join(map(str, [elm, spec, count, '%.10f' % frac]))+'\n')
