def buildContentSummary(categories, categoryData, database):
# builds the content summary for a database
iters = 2 if len(categories) > 1 else 1
keywords = [TAXONOMY.get(cat) for cat in categories[:iters]]
for i in range(iters):
keys = reduce(list.__add__, keywords[i:])
urls = getUniqueDocs(keys, categoryData)
logger("Building the content summary for " + categories[i] + \
". Total docs to fetch: " + str(len(urls)), highlight=True)
crawler.getContentSummary(database, categories[i], urls, categoryData)
def _sorted_homologenes(self, homologenes):
'''sort list of homologenes [(taxid, geneid),...] based on the order
defined in species_li.
'''
d = {}
for i, species in enumerate(list(TAXONOMY.keys())):
d[TAXONOMY[species]["tax_id"]] = i
gene_li = [(d.get(taxid, taxid), taxid, geneid)
for taxid, geneid in homologenes]
return [g[1:] for g in sorted(gene_li)]
def classifyDb(database, Tc=100, Ts=0.6):
# classifies a database based on values of
# threshold and specificity
categories, categoryData = ["Root"], {}
for cat in categories:
logger("Analyzing " + cat + " category")
filename = cat.lower() + ".txt"
keywords = TAXONOMY.get(cat)
if keywords:
queryUrlMap = buildQueryUrlMap(database, filename)
categoryData.update(queryUrlMap)
keywordCount = {k: sum([q["count"] for q in
queryUrlMap[k].itervalues()]) for k in keywords}
N = float(sum(keywordCount.values()))
for k, v in keywordCount.items():
logger("Coverage for {0} : {1}, Specificity: {2}".format(k, str(v), str(v/N)))
if v >= Tc and v/N >= Ts:
logger(">>>>>> Adding " + k + " to category <<<<<<")
categories.append(k)
return (categories, categoryData)
def __init__(self, data_folder):
# if species_li is None, include all species
self.set_species_li(list(TAXONOMY.keys()))
self.data_folder = data_folder
self.datafile = os.path.join(self.data_folder, self.DATAFILE)
