A simple script to traverse all GO classes

traverse_go_classes.py

import requests
import time
import pdb
import json
import sys
import urllib

FILE_NAME = "ontology_ids_and_parents.txt"
APIKEY = 'YOUR KEY'

def get_all_classes():
    more_pages = True
    payload = { 'apikey' : APIKEY , 'page' : 1,
                'include' : 'parents' }
    out = open(FILE_NAME,"w")
    while more_pages:
        print "Page %d"%payload['page'],
        sys.stdout.flush()
        r = requests.get("http://data.bioontology.org/ontologies/GO/classes", 
                params=payload)
        print " OK"
        data = json.loads(r.text)
        if 'collection' not in data:
            pdb.set_trace()
        for item in data['collection']:
            out.write(item["@id"])
            for parent in item['parents']:
                out.write("\t" + parent["@id"])
            out.write("\n")
        if 'nextPage' not in data:
            payload['page'] = None
        else:
            payload['page'] = data['nextPage']
        more_pages = payload['page'] != None
    out.close()

def test_coverage_parents():
    classes = set()
    parents = set()
    with open(FILE_NAME,"r") as fin:
        for line in fin:
            line = line.strip()
            data = line.split("\t")
            classes.add(data[0])
            for parent in data[1:]:
                parents.add(parent)
        fin.close()
    missing = parents.difference(classes)
    for cls in missing:
        print "Testing ", cls
        payload = { 'apikey' : APIKEY }
        cls = urllib.quote(cls,'')
        cls_call = "http://data.bioontology.org/ontologies/GO/classes/%s"%cls
        r = requests.get(cls_call,
                params=payload)
        time.sleep(1)
        print "\t",r.headers['status']
get_all_classes()
test_coverage_parents()