Source Code for Module TEES.Utils.Convert.SubtiWiki

 1  import sys, os 
 2  import codecs, time 
 3  from wikitools import wiki 
 4  from wikitools import api 
 5   
 6 -def loadPages(site, pages, outDir, wait=3): 
 7      count = 0 
 8      for page in pages: 
 9          print >> sys.stderr, "Loading page", page, "(" + str(count+1) +"/"+str(len(pages)) + ")" 
10          # define the params for the query 
11          params = {'action':'query', 'titles':page, 'export':None} 
12          # create the request object 
13          request = api.APIRequest(site, params) 
14          # query the API 
15          result = request.query() 
16           
17          print >> sys.stderr, "Writing result" 
18          f = codecs.open(os.path.join(outDir, page+".xml"), "wt", "utf-8") 
19          f.write( result["query"]["export"]["*"] ) 
20          f.close() 
21           
22          print >> sys.stderr, "Sleeping" 
23          time.sleep(wait) 
24          count += 1 
25       
26 -def readProteinNames(file): 
27      f = codecs.open(file, "rt", "utf-8") 
28      names = [] 
29      for line in f: 
30          names.append(line.strip()) 
31      return names 
32   
33 -def getSynonyms(inDir, outFilename): 
34      outFile = open(outFilename, "wt") 
35      for file in sorted(os.listdir(inDir)): 
36          if file.find("xml") != -1: 
37              protName = file.split(".")[0] 
38              outFile.write(protName) 
39              f = codecs.open(os.path.join(inDir, file), "rt", "utf-8") 
40              geneNameLine = False 
41              for line in f: 
42                  if geneNameLine: 
43                      if line.find(protName) == -1: 
44                          print "Warning,", protName, "not found" 
45                          geneNameLine = False 
46                  if line.find("||") == -1: # not table line 
47                      continue 
48                  if line.find("Gene Name") != -1: 
49                      geneNameLine = True 
50                  elif line.find("Synonyms") != -1: 
51                      synString = line.split("||")[1] 
52                      synString = synString.replace("''", "") 
53                      synonyms = synString.split(",") 
54                      for i in range(len(synonyms)): 
55                          #synonyms[i] = synonyms[i].strip() 
56                          synonym = synonyms[i].strip() 
57                          if synonym != "": 
58                              outFile.write(","+synonym) 
59              outFile.write("\n")              
60   
61  if __name__=="__main__": 
62      import sys 
63      from optparse import OptionParser 
64      # Import Psyco if available 
65      try: 
66          import psyco 
67          psyco.full() 
68          print >> sys.stderr, "Found Psyco, using" 
69      except ImportError: 
70          print >> sys.stderr, "Psyco not installed" 
71   
72      optparser = OptionParser(description="Extract information from SubtiWiki for the BioNLP'11 REN task") 
73      optparser.add_option("-i", "--input", default="http://subtiwiki.uni-goettingen.de/wiki/api.php", dest="input", help="") 
74      optparser.add_option("-o", "--output", default="subtiwiki/pages", dest="output", help="") 
75      optparser.add_option("-n", "--names", default="subtiwiki/Subtiwiki-Protein-Coding-Genes-Names.txt", dest="names", help="") 
76      optparser.add_option("-s", "--synonyms", default="subtiwiki/Subtiwiki-Synonyms.csv", dest="synonyms", help="") 
77      (options, args) = optparser.parse_args() 
78   
79      # create a Wiki object 
80      site = wiki.Wiki(options.input) 
81      loadPages(site, readProteinNames(options.names), options.output, wait=5) 
82      getSynonyms(options.output, options.synonyms) 
83