[WIP] added scripts to extract owlsim data from biolink

scripts/golr-exporter.py

@@ -0,0 +1,91 @@
+#!/usr/bin/python
+
+import os
+import getopt
+import urllib
+import json
+import csv
+import shutil
+import argparse
+
+def uniqAndSort(output):
+    sorted_suffix = '.sorted2'
+    f = open(output, "r")
+    fs = open(output + sorted_suffix, "w")
+    fs.writelines(sorted(set(f.readlines())))
+    shutil.move(output + sorted_suffix, output)
+
+def transformLabel(input, output):
+    with open(input) as data_file:
+        data = json.load(data_file)
+
+    with open(output, 'w') as tsvfile:
+        writer = csv.writer(tsvfile, delimiter='\t')
+        for entry in data:
+            id = entry["subject"]
+            label = entry["subject_label"]
+            writer.writerow([id, label])
+
+    uniqAndSort(output)
+
+def transformAssociation(input, output):
+    with open(input) as data_file:
+        data = json.load(data_file)
+
+    with open(output, 'w') as tsvfile:
+        writer = csv.writer(tsvfile, delimiter='\t')
+        for entry in data:
+            id = entry["subject"]
+            for obj in entry["objects"]:
+                writer.writerow([id, obj])
+
+    uniqAndSort(output)
+
+def main():
+
+    biolink = "http://localhost:5000/api" #mart/labels/gene/phenotype/NCBITaxon%3A7955
+
+    taxon_map = {
+        'Hs': 9606,
+        'Mm': 10090,
+        'Dr': 7955,
+        'Dm': 7227,
+        'Ce': 6239
+        }
+
+    parser = argparse.ArgumentParser(description='Fetcher from biolink for monarch data',
+                                     formatter_class=argparse.RawTextHelpFormatter)
+    parser.add_argument('-t', '--taxon', type=str, required=False,
+                        help='species prefix: ' + ",".join(taxon_map.keys()))
+    args = parser.parse_args()
+
+    if args.taxon is None:
+        tax_list = taxon_map.keys()
+    else:
+        tax_list = args.taxon.split(',')
+
+    print("Running for: " + ",".join(tax_list))
+    for tax in tax_list:
+        if not os.path.exists(tax):
+            os.makedirs(tax)
+
+        subjs = ["gene"]
+        obj = "phenotype"
+
+        if tax == "Hs":
+            subjs = ["disease", "case"]
+
+        for subj in subjs:
+            assocFileJson = tax + "/" + subj + "-" + obj + ".json"
+            assocFileTsv = tax + "/" + subj + "-" + obj + ".tsv"
+            labelFileTsv = tax + "/" + subj + "-label.tsv"
+
+            assocURL = biolink + "/mart/" + subj + "/" + obj + "/NCBITaxon:" + str(taxon_map.get(tax))
+            print("fetching " + assocURL)
+            urllib.urlretrieve (assocURL, assocFileJson)
+            transformAssociation(assocFileJson, assocFileTsv)
+            transformLabel(assocFileJson, labelFileTsv)
+            os.remove(assocFileJson)
+
+if __name__ == "__main__":
+    main()