def parseInteractors(dom, network): ems = dom.getElementsByTagName("interactor") interactors = [] #print "interactor tags %d " %len(ems) for em in ems: interactor = Entity() tmp = em.getElementsByTagName("shortLabel") slabel = tmp[0].firstChild.nodeValue.strip() slabel = slabel.split('_')[0] interactor.symbol = slabel interactor.refs = {} interactor.name = getText(em, 'fullName') setPrimaryRef(em, interactor) if 'uniprotkb' in interactor.refs: interactor._id = "enti_up_%s" % (interactor.refs['uniprotkb']) else: interactor._id = idtool.generate("entity") interactor.refs['intact'] = em.getAttribute('id') interactor.group = getText( em.getElementsByTagName("interactorType")[0], 'shortLabel') interactors.append(interactor) network.entities = interactors
def parseInteractors(dom, network): ems = dom.getElementsByTagName("interactor") interactors = [] #print "interactor tags %d " %len(ems) for em in ems: interactor = Entity() tmp = em.getElementsByTagName("shortLabel") slabel = tmp[0].firstChild.nodeValue.strip() slabel = slabel.split('_')[0] interactor.symbol = slabel interactor.refs = {} interactor.name = getText(em, 'fullName') setPrimaryRef(em, interactor) if 'uniprotkb' in interactor.refs: interactor._id = "enti_up_%s" %(interactor.refs['uniprotkb']) else: interactor._id = idtool.generate("entity") interactor.refs['intact'] = em.getAttribute('id') interactor.group = getText( em.getElementsByTagName("interactorType")[0], 'shortLabel') interactors.append(interactor) network.entities = interactors
def importEdges(filename=None): """ import edges, one pair per line Edge property: - no direction, -> left to right, <- right to left """ filename = filename or r'C:\work\caida\Dropbox\precon\engineering\Contents\GBM_BN-Massaged.csv' # add a network n = Network.findOne({'name': "GBM Predicted Tumor Network"}) or Network() n.name = "GBM Predicted Tumor Network" n.group = "predicted" n.owner = 'precon' n.save() col = mongo.db()['entity'] count = 0 now = time.time() ec = mongo.getCollection('edge') with open(filename) as f: content = f.read() for line in content.split('\n'): doc = {} line = line.strip() pos = line.find("///") if pos > 0: line = line[0:pos].strip() doc['comment'] = line[pos:] items = line.split("->") if (len(items) == 1): items = line.split("<-") if (len(items) == 1): items = line.split(" - ") if (len(items) == 1): error("Ignore invalid line: [%s]" % line) continue count += 1 if (count < 8378): continue tmp = [] tmp.append(items[0].lower().strip()) tmp.append(items[1].lower().strip()) entities = ['', ''] print "!!! %d " % (col.find({'symbol': {'$in': tmp}}).count()) for r in col.find({'symbol': {'$in': tmp}}): if (r['symbol'] == tmp[0]): entities[0] = Entity(r) if (r['symbol'] == tmp[1]): entities[1] = Entity(r) if (len(entities) != 2): raise "Invalid nodes %s, continue" % entities node1 = Node.findOne({'network': n._id, 'entity': entities[0]._id}) if not node1: node1 = Node({'network': n._id}, entities[0]) node1.save() node2 = Node.findOne({'network': n._id, 'entity': entities[1]._id}) if not node2: node2 = Node({'network': n._id}, entities[1]) node2.save() con = Connection() con._id = "conn_%s_%s" % (tmp[0], tmp[1]) con.nodes = [node1._id, node2._id] con.entities = [entities[0]._id, entities[1]._id] con.type = 'predicted' con.network = n._id con.label = "" con.save() print "Saving connection %d %s" % (count, con._id) finish = time.time() print "Imported %d edges, time elpased %d seconds" % (count, finish - now)
def load(self): self.columnIndexes = {} headers = self.rows.next() for i in range (len(headers)): col = headers[i] if col not in SUPPORTED_COLUMNS: raise Exception("Invalid column: %s. Accepted columns are: %s" %(col, SUPPORTED_COLUMNS)) self.columnIndexes[col] = i #networks, connections, entities, nodes entities = {} nodes = {} networks = {} for row in self.rows: entityA = None dbid = self.getCol('NodeA_DBID',row ) if dbid: dbid = dbid.lower() _id = 'enti_%s' %dbid if _id not in entities: entityA = Entity() entityA._id = _id entityA.dbref = {} pair = dbid.split(':') if(len(pair)!=2): self._error("Invalid NodeA_DBID: %s" %dbid) continue entityA.dbref[pair[0]] = pair[1] entityA.group = self.getCol('NodeA_Category',row) #entity.label = 'Metformin' entityA.name = self.getCol('NodeA',row) entities[_id] = entityA else: entityA=entities[_id] #Node B entityB = None dbid = self.getCol('NodeB_DBID',row, False) if dbid: dbid = dbid.lower() _id = 'enti_%s' %dbid if _id not in entities: entityB = Entity() entityB._id = _id entityB.dbref = {} pair = dbid.split(':') if(len(pair)!=2): self._error("Invalid NodeB_DBID: %s" %dbid) continue entityB.dbref[pair[0].strip()] = pair[1].strip() entityB.group = self.getCol('NodeB_Category',row) #entity.label = 'Metformin' entityB.name = self.getCol('NodeB' ,row ) entities[_id] = entityB else: entityB = entities[_id] edgeType = self.getCol('Edge',row, False) if not edgeType: continue con = Connection() con.type = edgeType nodeA = nodes[entityA._id] if entityA._id in nodes else self.newNode(entityA) nodeB = nodes[entityB._id] if entityB._id in nodes else self.newNode(entityB) nodes[entityA._id] = nodeA nodes[entityB._id] = nodeB con.entities=[entityA._id, entityB._id] # this is for search con.nodes = [nodeA._id, nodeB._id] con._nodes = [nodeA, nodeB] con.refs = {} edgeRefs = self.getCol('Edge_Ref',row, False) if edgeRefs: # comma separated pairs = edgeRefs.split(",") if len(pairs) == 1: pairs = edgeRefs.split("\n") for p in pairs: pair = p.split(':') refs = pair[1].split(';') if len(refs) == 1: refs = pair[1] con.refs[pair[0].lower().strip()] = refs networkName = self.getCol('Network',row) or 'DEFAULT' network = networks[networkName] if networkName in networks else Network() network.name = networkName network.owner='peop_precon' networks[networkName] = network network._connections = network._connections or [] network._connections.append(con) con.network = network._id con.owner='precon' #TBD, error return networks, self.errors