def getSynIDs(oboLocation, selOboIDs):
ontology = GeneOntology(oboLocation)
allSynIDs = []
for termID in ontology.dTerms:
oboNode = ontology.dTerms[termID]
oboID = oboNode.id
oboName = oboNode.name
oboSyns = oboNode.synonym
oboRels = oboNode.is_a
if oboID in selOboIDs:
allSynIDs.append(oboID)
allchildren = oboNode.getAllChildren()
allSynIDs += [x.term.id for x in allchildren]
allSynIDs = [x.replace(":", '_') for x in allSynIDs]
return allSynIDs
def printTerms(oboLocation, selOboIDs, excludeIDs=None, printDepht=3):
ontology = GeneOntology(oboLocation)
allExclIDs = set()
if excludeIDs != None:
for elem in excludeIDs:
exclElem = ontology.dTerms[elem]
allChildren = exclElem.getAllChildren()
for child in allChildren:
allExclIDs.add(child.term.id)
for termID in ontology.dTerms:
oboNode = ontology.dTerms[termID]
oboID = oboNode.id
oboName = oboNode.name
oboSyns = oboNode.synonym
oboRels = oboNode.is_a
if oboID in selOboIDs:
print("Base", oboID, oboName)
printChildren(oboNode, 5, 1, allExclIDs)
dbData = DataFrame.parseFromFile(miRExploreDir + "/miR2Disease/mirna_disease.tsv", bConvertTextToNumber=False)
allDiseases = set()
for row in dbData:
disease = row['disease']
if disease == 'None':
continue
allDiseases.add(disease.upper())
print(len(allDiseases))
diseaseObo = GeneOntology(miRExploreDir + "/doid.obo")
disease2obo = defaultdict(set)
"""
find perfect matches
"""
for oboID in diseaseObo.dTerms:
oboNode = diseaseObo.dTerms[oboID]
doidName = oboNode.name
upperDiseaseName = doidName.upper()
if upperDiseaseName in allDiseases:
args = parser.parse_args()
#resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
resultBase = args.resultdir
diseaseSyns = SynfileMap(resultBase + "/go/synfile.map")
diseaseSyns.loadSynFiles(
('/home/users/joppich/ownCloud/data/', args.datadir))
allfiles = glob.glob(resultBase + "/go/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
fmaObo = GeneOntology(args.datadir + "miRExplore/go/go.obo")
def getTerm(synid, obo):
if synid in obo.dTerms:
return obo.getID(synid)
synid = synid.replace('_', ':', 1)
return obo.getID(synid)
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
dataDir
from database.Neo4JInterface import neo4jInterface
from utils.parallel import MapReduce
from enum import Enum
resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
diseaseSyns = SynfileMap(resultBase + "/disease/synfile.map")
diseaseSyns.loadSynFiles(('/home/users/joppich/ownCloud/data/', dataDir))
allfiles = glob.glob(resultBase + "/disease/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
celloObo = GeneOntology(dataDir + "miRExplore/doid.obo")
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
diseaseFile = resultBase + "/disease/" + splitFileID + ".index"
diseaseHits = SyngrepHitFile(diseaseFile, diseaseSyns)
if len(diseaseHits) == 0:
continue
sentFile = "/mnt/c/dev/data/pmc/allsent/" + splitFileID + ".sent"
dataDir
from database.Neo4JInterface import neo4jInterface
from utils.parallel import MapReduce
from enum import Enum
resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
diseaseSyns = SynfileMap(resultBase + "/cellline/synfile.map")
diseaseSyns.loadSynFiles(('/home/users/joppich/ownCloud/data/', dataDir))
allfiles = glob.glob(resultBase + "/cellline/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
celloObo = GeneOntology(dataDir + "miRExplore/meta_cells.obo")
def getTerm(synid, obo):
if synid in obo.dTerms:
return obo.getID(synid)
synid = synid.replace('_', ':', 1)
return obo.getID(synid)
def analyseFile(splitFileIDs, env):
fileCoocs = []
dataDir
from database.Neo4JInterface import neo4jInterface
from utils.parallel import MapReduce
from enum import Enum
resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
diseaseSyns = SynfileMap(resultBase + "/go/synfile.map")
diseaseSyns.loadSynFiles(('/home/users/joppich/ownCloud/data/', dataDir))
allfiles = glob.glob(resultBase + "/go/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
fmaObo = GeneOntology(dataDir + "miRExplore/go/go.obo")
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
diseaseFile = resultBase + "/go/" + splitFileID + ".index"
diseaseHits = SyngrepHitFile(diseaseFile, diseaseSyns)
if len(diseaseHits) == 0:
continue
sentFile = "/mnt/c/dev/data/pmc/allsent/" + splitFileID + ".sent"
args = parser.parse_args()
#resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
resultBase = args.resultdir
diseaseSyns = SynfileMap(resultBase + "/cellline/synfile.map")
diseaseSyns.loadSynFiles(
('/home/users/joppich/ownCloud/data/', args.datadir))
allfiles = glob.glob(resultBase + "/cellline/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
celloObo = GeneOntology(args.datadir + "miRExplore/meta_cells.obo")
def getTerm(synid, obo):
if synid in obo.dTerms:
return obo.getID(synid)
synid = synid.replace('_', ':', 1)
return obo.getID(synid)
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
from collections import defaultdict
from nertoolkit.geneontology.GeneOntology import GeneOntology
from synonymes.Synonym import Synonym
from synonymes.SynonymUtils import handleCommonExcludeWords
from utils.idutils import dataDir, loadExludeWords, printToFile, speciesName2TaxID
from collections import defaultdict
from nertoolkit.geneontology.GeneOntology import GeneOntology
from synonymes.Synonym import Synonym
from synonymes.SynonymUtils import handleCommonExcludeWords
from utils.idutils import dataDir, loadExludeWords, printToFile, speciesName2TaxID
celloObo = GeneOntology(dataDir + "miRExplore/cell_ontology/cl.obo")
vAllSyns = []
allOboNames = defaultdict(set)
for cellID in celloObo.dTerms:
oboNode = celloObo.dTerms[cellID]
oboID = oboNode.id
if not oboID.startswith("CL"):
continue
oboName = oboNode.name
allOboNames[oboName].add(oboID)
taxName = " ".join(aTaxName[0:min(2, len(aTaxName))])
try:
int(taxName)
except ValueError:
globalKeywordExcludes.add(taxName)
globalKeywordExcludes = sorted(globalKeywordExcludes)
printToFile(
globalKeywordExcludes,
dataDir + "miRExplore/textmine/excludes/exclude_words.names.syn")
print("Done: names")
if addGO:
goObo = GeneOntology(dataDir + "miRExplore/textmine/excludes/go.obo")
globalKeywordExcludes = set()
for id in goObo.dTerms:
child = goObo.dTerms[id]
if 'cellular_component' in child.namespace:
globalKeywordExcludes.add(child.name)
if child.synonym != None:
for syn in child.synonym:
if syn != None:
synWord = syn.syn
args = parser.parse_args()
#resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
resultBase = args.resultdir
diseaseSyns = SynfileMap(resultBase + "/disease/synfile.map")
diseaseSyns.loadSynFiles(
('/home/users/joppich/ownCloud/data/', args.datadir))
allfiles = glob.glob(resultBase + "/disease/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
celloObo = GeneOntology(args.datadir + "/miRExplore/doid.obo")
def getTerm(synid, obo):
if synid in obo.dTerms:
return obo.getID(synid)
synid = synid.replace('_', ':', 1)
return obo.getID(synid)
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
from collections import defaultdict
from nertoolkit.geneontology.GeneOntology import GeneOntology
from database.Neo4JInterface import neo4jInterface
from synonymes.Synonym import Synonym
from synonymes.SynonymUtils import handleCommonExcludeWords
from utils.idutils import dataDir, loadExludeWords, printToFile, speciesName2TaxID, eprint
celloObo = GeneOntology(dataDir + "miRExplore/cellosaurus/cellosaurus.obo")
tax2cells = defaultdict(set)
id2node = {}
id2species = defaultdict(set)
id2derived_from = defaultdict(set)
allowedTaxIDs = set([str(speciesName2TaxID[x]) for x in speciesName2TaxID])
for cellID in celloObo.dTerms:
oboNode = celloObo.dTerms[cellID]
oboID = oboNode.id
oboName = oboNode.name
oboSyns = oboNode.synonym
oboXRefs = oboNode.xref
oboRels = oboNode.is_a
taxID = set()
if oboXRefs != None:
for xref in oboXRefs:
seenElements[iprElem.id] = iprElem
noParents = 0
for iprID in seenElements:
iprElem = seenElements[iprID]
if len(iprElem.parents) == 0:
noParents += 1
for id in iprElem.parents:
if not id in seenElements:
print(id)
print(iprElem)
print("Elem Count", len(seenElements))
print("Root Count", noParents)
interProObo = GeneOntology()
for iprID in seenElements:
iprElem = seenElements[iprID]
iprGoElem = iprElem.to_go_term()
interProObo.dTerms[iprID] = iprGoElem
interProObo.linkChildren()
interProObo.saveFile(interproFolder + "/interpro.obo")
from collections import defaultdict
from nertoolkit.geneontology.GeneOntology import GeneOntology
from database.Neo4JInterface import neo4jInterface
from synonymes.Synonym import Synonym
from synonymes.SynonymUtils import handleCommonExcludeWords
from utils.idutils import dataDir, loadExludeWords, printToFile, speciesName2TaxID
diseaseObo = GeneOntology(dataDir + "miRExplore/doid.obo")
tax2cells = defaultdict(set)
id2node = {}
id2derived_from = defaultdict(set)
for cellID in diseaseObo.dTerms:
oboNode = diseaseObo.dTerms[cellID]
oboID = oboNode.id
oboName = oboNode.name
oboRels = oboNode.is_a
id2node[oboID] = {'id': oboID, 'name': oboName}
if oboRels != None:
for rel in oboRels:
term = rel.term
id2derived_from[oboID].add(term.id)
db = neo4jInterface(simulate=False)
if len(line) > 0:
accept_pmids.add(line)
#resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
resultBase = args.resultdir
oboSyns = SynfileMap(resultBase + "/synfile.map")
oboSyns.loadSynFiles(('/home/users/joppich/ownCloud/data/', args.datadir))
allfiles = glob.glob(resultBase + "/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
celloObo = GeneOntology(args.obo.name)
def getTerm(synid, obo):
if synid in obo.dTerms:
return obo.getID(synid)
synid = synid.replace('_', ':', 1)
return obo.getID(synid)
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
args = parser.parse_args()
#resultBase = dataDir + "/miRExplore/textmine/results_pmc/"
resultBase = args.resultdir
diseaseSyns = SynfileMap(resultBase + "/model_anatomy/synfile.map")
diseaseSyns.loadSynFiles(
('/home/users/joppich/ownCloud/data/', args.datadir))
allfiles = glob.glob(resultBase + "/model_anatomy/*.index")
allfileIDs = [os.path.basename(x).replace(".index", "") for x in allfiles]
allfileIDs = sorted(allfileIDs, reverse=True)
#allfileIDs = [894]
fmaObo = GeneOntology(args.datadir +
"miRExplore/foundational_model_anatomy/fma_obo.obo")
def getTerm(synid, obo):
if synid in obo.dTerms:
return obo.getID(synid)
synid = synid.replace('_', ':', 1)
return obo.getID(synid)
def analyseFile(splitFileIDs, env):
fileCoocs = []
for splitFileID in splitFileIDs:
from collections import defaultdict
from nertoolkit.geneontology.GeneOntology import GeneOntology
from synonymes.Synonym import Synonym
from synonymes.SynonymUtils import handleCommonExcludeWords
from utils.idutils import dataDir, loadExludeWords, printToFile, speciesName2TaxID
celloObo = GeneOntology(dataDir + "miRExplore/textmine/neutrophils.obo")
vAllSyns = []
for cellID in celloObo.dTerms:
oboNode = celloObo.dTerms[cellID]
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)
#print(str(taxID) + " " + str(newSyn))
print(newSyn)