ncitObo = GeneOntology(dataDir + "miRExplore/obodir/ncit.obo")
ncitTerm2Sym = NcitTermSymbolDB.loadFromFolder()
vAllSyns = []
for termID in ncitObo.dTerms:
oboNode = ncitObo.dTerms[termID]
oboID = oboNode.id
oboName = oboNode.name
oboSyns = oboNode.synonym
oboRels = oboNode.is_a
newSyn = Synonym(oboID)
newSyn.addSyn(oboName)
if oboSyns != None:
for x in oboSyns:
newSyn.addSyn(x.syn)
allOrgs = [x for x in ncitTerm2Sym.org_term2symbol]
for org in allOrgs:
ncitID = oboID[oboID.index(":") + 1:]
if ncitID in ncitTerm2Sym.org_term2symbol[org]:
orgSyms = ncitTerm2Sym.org_term2symbol[org][ncitID]
if x.startswith("EC "):
continue
if x.upper() in ["TH1"]:
continue
allSyms.append(x)
printID = mgiID.replace(':', '_', 1)
if printID in locID2sym and len(locID2sym[printID]) > 0:
allSyms.append(mgiID)
allSyms.append(mgiID.replace(':', '_'))
printID = locID2sym[printID]
synline = printID + ":" + "|".join(allSyms)
synonyme = Synonym.parseFromLine(synline)
vAllSyns.append(synonyme)
for syn in vAllSyns:
removeSyns = []
for synword in syn.syns:
if len(synword) == 1:
removeSyns.append(synword)
if len(removeSyns) > 0:
print(syn.id, removeSyns)
syn.removeSyn(removeSyns)
bodypartsObo = GeneOntology(
dataDir + "miRExplore/foundational_model_anatomy/fma_obo.obo")
vAllSyns = []
for cellID in bodypartsObo.dTerms:
oboNode = bodypartsObo.dTerms[cellID]
oboID = oboNode.id
oboName = oboNode.name
oboSyns = oboNode.synonym
oboRels = oboNode.is_a
newSyn = Synonym(oboID)
newSyn.addSyn(oboName)
aName = oboName.split(' ')
if len(aName) > 1 and len(aName) < 5:
acro = ""
if aName[-1].upper() == 'CELL':
acro = "".join([x[0].upper() for x in aName])
newSyn.addSyn(acro)
if oboSyns != None:
for x in oboSyns:
newSyn.addSyn(x.syn)
aline = [x.strip() for x in line.split('\t')]
name = aline[1]
altNames = StringIO()
altNames.write(aline[2] + "\n")
names = []
for line in csv.reader([aline[2]], dialect='phenotypes'):
for elem in line:
names.append(elem)
names = names + [name]
newSyn = Synonym( 'DISEASE' + str(len(vAllSyns)+1))
newSyn.addSyn(name)
for x in names:
if x.startswith('[D]') or x.startswith('[X]') or x.startswith('[M]'):
x = x[3:]
xsyns = []
if ' - ' in x:
xsyns += x.split(' - ')
else:
xsyns.append(x)
for xsyn in xsyns:
oboNode = celloObo.dTerms[cellID]
oboID = oboNode.id
if not oboID.startswith("CL"):
continue
if oboID == 'CL:1000413':
print(oboID)
print(oboNode.name)
oboName = oboNode.name
oboSyns = oboNode.synonym
oboRels = oboNode.is_a
newSyn = Synonym(oboID)
newSyn.addSyn(oboName)
if oboSyns != None:
for x in oboSyns:
if x == None:
continue
if x.syn in allOboNames:
continue
newSyn.addSyn(x.syn)
for x in newSyn.syns:
allNodes.append(oboNode)
globalKeywordExcludes = loadExludeWords(common=False,
cell_co=False,
disease=False,
generic=False)
for x in globalKeywordExcludes:
if 'membrane' in globalKeywordExcludes[x]:
print("Membrane: " + x)
synSet = set()
for node in allNodes:
newSyn = Synonym(node.id)
newSyn.addSyn(node.name)
if node.synonym != None:
for x in node.synonym:
if x == None:
continue
newSyn.addSyn(x.syn)
synSet.add(newSyn)
vPrintSyns = handleCommonExcludeWords(synSet,
globalKeywordExcludes,
mostCommonCount=66,
maxCommonCount=5,
minSynCount=0)