def run_zscore(gse, indir, outdir, probe_sets):
start = time.time()
path = os.path.join(indir, gse[0], gse[1] + '*.csv')
probe_set_files = [os.path.join(probe_set_file_location, i, gse[0] + '-' + i.lower().replace('-', '_') + '.csv') for i in probe_sets]
outpath = os.path.join(outdir, gse[0])
make_sure_path_exists(outpath)
gse_file = glob.glob(path)
if len(gse_file) == 1:
zscores.script_run(
gse_file[0],
gse[2],
gse[3],
metadata_feature_rows=gse[4:],
probe_set_files=probe_set_files,
outdir=outpath)
print 'Finished', gse[1], ', took', time.time() - start, 's'
def do_threaded_work(multivariates):
for i, col in multivariates.iteritems():
gpl = col.loc['GPL']
gse = col.loc['GSE'].upper()
col = col.dropna()
path = os.path.join('multivariate-clinical', 'Microarray-datasets', gpl, '*' + gse + '*.csv')
fnames = glob.glob(path)
print fnames
if len(fnames) != 1:
print "Failed to find file for: ", gpl, gse
sys.exit(1)
fname = fnames[0]
feature_names = make_feature_names(col)
out_path = os.path.join('multivariate-out', gpl)
make_sure_path_exists(out_path)
if len(glob.glob(os.path.join(out_path, gse + '*.csv'))) >= 1:
continue
zscores.script_run(fname,
col.loc['Time'],
col.loc['Censor'],
metadata_feature_rows=feature_names,
outdir=out_path)
