view xenaGetDataset/xenaGetDataset.py @ 0:aa2ed7c6c09b

Uploaded

#!/usr/bin/env python

import argparse
import json
import urllib2
import re
from functools import partial

# import xena query utilities
import xena_query as xena

# xena url for public datasets, at ucsc
xena_url = "https://genome-cancer.ucsc.edu/proj/public/xena"

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("datasetType", type=str)
    parser.add_argument("cohortName", type=str)
    parser.add_argument("datasetName", type=str)
    parser.add_argument("outfile", type=str)
    args = parser.parse_args()

    # short-hand wrappers for xena methods
    post = xena.compose(json.loads, partial(xena.post, xena_url))
    find_datasets_type_pattern = xena.compose(post, 
					      xena.find_datasets_type_pattern)
    name_to_url = partial(xena.name_to_url, xena_url)

# Find all TCGA mutation calls by matching on the type and dataset name.
    fullDatasetName = "%s/%s" % (args.cohortName, args.datasetName)
    fullDatasetName = "%" + fullDatasetName
    matchingDatasets = find_datasets_type_pattern(args.datasetType, 
                                                  fullDatasetName)
    if len(matchingDatasets) > 1:
        errorMsg = "Error: matches to multiple datasets:"
	for dataset in matchingDatasets:
            errorMsg = "%s %s" % (errorMsg, dataset)
	exit(errorMsg)
    else:
        ff = urllib2.urlopen(name_to_url(matchingDatasets[0]))
	with open(args.outfile, "w") as out:
            out.write(ff.read())


if __name__ == "__main__":
    main()