def flattenForClinicalFeature(sMap, bookDic):
clinFeatures = []
finalClinFeature = None
sampleMap = sMap.getName()
datasets = collectNamesBelongToSampleMap(bookDic, sampleMap)
for name in datasets:
obj = bookDic[name]
if obj['type'] == "clinicalMatrix":
clinFeature = None
#clinFeature obj
if obj.has_key(':clinicalFeature'):
path = bookDic[obj[':clinicalFeature']]['path']
neme = bookDic[obj[':clinicalFeature']]['name']
clinFeature = ClinicalFeatureNew(path, name)
#get matrix obj
path = obj['path']
name = obj['name']
cMatrix = ClinicalMatrixNew(path, name, False, clinFeature)
if clinFeature:
clinFeatures.append(clinFeature)
fout = open(".tmp", 'w')
fout.close()
for clinF in clinFeatures:
fout = open(".tmptmp", 'w')
clinF.store(fout)
fout.close()
os.system("cat .tmptmp >> .tmp")
fin = open(".tmp", 'r')
jsonName = trackName_fix(sampleMapBaseName(sMap) + "_clinicalFeature")
finalClinFeature = ClinicalFeatureNew(fin, jsonName)
if not finalClinFeature.isValid():
print "final clinFeature file .tmp is invalid"
return 0
fin.close()
#vis exceptions
VIS_limit = 4
if bookDic.has_key(
sMap.getName()) and bookDic[sMap.getName()].has_key("VIS"):
VIS_limit = bookDic[sMap.getName()]["VIS"]
finalClinFeature.fillInPriorityVisibility(VIS_limit)
finalClinFeature.setFeatureShortTitle("_PATIENT", "_PATIENT_ID")
finalClinFeature.setFeatureLongTitle("_PATIENT", "_PATIENT_ID")
finalClinFeature.setFeatureValueType("_PATIENT", "category")
finalClinFeature.setFeatureShortTitle("_INTEGRATION", "_SAMPLE_ID")
finalClinFeature.setFeatureLongTitle("_INTEGRATION", "_SAMPLE_ID")
finalClinFeature.setFeatureValueType("_INTEGRATION", "category")
return finalClinFeature
def flattenForClinicalFeature(sMap, bookDic):
clinFeatures=[]
finalClinFeature=None
sampleMap = sMap.getName()
datasets = collectNamesBelongToSampleMap(bookDic, sampleMap)
for name in datasets:
obj= bookDic[name]
if obj['type']=="clinicalMatrix":
clinFeature=None
#clinFeature obj
if obj.has_key(':clinicalFeature'):
path= bookDic[obj[':clinicalFeature']]['path']
neme = bookDic[obj[':clinicalFeature']]['name']
clinFeature = ClinicalFeatureNew(path, name)
#get matrix obj
path = obj['path']
name = obj['name']
cMatrix = ClinicalMatrixNew(path,name,False, clinFeature)
if clinFeature:
clinFeatures.append(clinFeature)
fout = open(".tmp",'w')
fout.close()
for clinF in clinFeatures:
fout = open(".tmptmp",'w')
clinF.store(fout)
fout.close()
os.system("cat .tmptmp >> .tmp")
fin = open(".tmp",'r')
jsonName= trackName_fix(sampleMapBaseName(sMap)+"_clinicalFeature")
finalClinFeature =ClinicalFeatureNew(fin,jsonName)
if not finalClinFeature.isValid():
print "final clinFeature file .tmp is invalid"
return 0
fin.close()
#vis exceptions
VIS_limit=4
if bookDic.has_key(sMap.getName()) and bookDic[sMap.getName()].has_key("VIS"):
VIS_limit= bookDic[sMap.getName()]["VIS"]
finalClinFeature.fillInPriorityVisibility(VIS_limit)
finalClinFeature.setFeatureShortTitle("_PATIENT","_PATIENT_ID")
finalClinFeature.setFeatureLongTitle("_PATIENT","_PATIENT_ID")
finalClinFeature.setFeatureValueType("_PATIENT","category")
finalClinFeature.setFeatureShortTitle("_INTEGRATION","_SAMPLE_ID")
finalClinFeature.setFeatureLongTitle("_INTEGRATION","_SAMPLE_ID")
finalClinFeature.setFeatureValueType("_INTEGRATION","category")
return finalClinFeature
def flattenEachSampleMap(sMap, bookDic,onlyGenomicSamples):
sampleMap = sMap.getName()
jsonName= trackName_fix(sampleMapBaseName(sMap)+"_clinicalMatrix")
finalClinMatrix= ClinicalMatrixNew(None,jsonName)
finalClinMatrixJSON={}
finalClinMatrixJSON["name"]=jsonName
finalClinMatrixJSON["type"]="clinicalMatrix"
finalClinMatrixJSON["path"]=""
finalClinMatrixJSON[":sampleMap"]=sampleMap
clinFeatures=[]
finalClinFeatureJSON=None
finalClinFeature=None
# add all ids to sMap
sMapChanged= checkIdsAllIn(sMap, bookDic)
#build initial clinical Matrix with sampleMap ids, all with empty data
emptyData={}
success = finalClinMatrix.addNewRows(sMap.getNodes(),emptyData)
if not success:
print "fail to add all initial ids from sampleMap"
datasets = collectNamesBelongToSampleMap(bookDic, sampleMap)
datasetsOrdered =[] #only the ClinicalMatrix ordered list
for name in datasets:
obj= bookDic[name]
if obj['type']=="clinicalMatrix":
if obj.has_key('outOfDate') and obj['outOfDate'] in ["yes", "Yes","YES"]:
datasetsOrdered.append(name)
elif not obj.has_key('outOfDate') and not obj.has_key('upToDate'):
datasetsOrdered.insert(0,name)
upToDateSets={}
for name in datasets:
obj= bookDic[name]
if obj['type']=="clinicalMatrix":
if obj.has_key('upToDate') :
upToDateSets[obj['upToDate']]=name
keys= upToDateSets.keys()
keys.sort()
for version in keys:
name = upToDateSets [version]
datasetsOrdered.insert(0,name)
for name in datasetsOrdered:
obj= bookDic[name]
if obj['type']=="clinicalMatrix":
clinFeature=None
#clinFeature obj
if obj.has_key(':clinicalFeature'):
path= bookDic[obj[':clinicalFeature']]['path']
neme = bookDic[obj[':clinicalFeature']]['name']
clinFeature = ClinicalFeatureNew(path, name)
#get matrix obj
path = obj['path']
name = obj['name']
cMatrix = ClinicalMatrixNew(path,name,False, clinFeature)
if finalClinMatrix==None:
finalClinMatrix= cMatrix
if finalClinMatrixJSON==None:
finalClinMatrixJSON= obj
#merge final and cMatrix
if finalClinMatrix != cMatrix:
print "name=",cMatrix.getName()
r = finalClinMatrix.addNewCols(cMatrix,validation=True)
if r!=True:
print "Fail to merge"
return False
#add clinFeature
if clinFeature:
clinFeatures.append(clinFeature)
#merge finalClinMatrixJSON with new json
if finalClinMatrixJSON != obj:
jsonName= trackName_fix(sampleMapBaseName(sMap)+"_clinicalMatrix")
finalClinMatrixJSON= cgDataMergeJSON(finalClinMatrixJSON, obj, jsonName)
# final ClinFeature json
if clinFeature:
clinFeatureJSON = bookDic[obj[':clinicalFeature']]
if finalClinFeatureJSON==None:
finalClinFeatureJSON= clinFeatureJSON
else:
jsonName= trackName_fix(sampleMapBaseName(sMap)+"_clinicalFeature")
finalClinFeatureJSON["version"]=datetime.date.today().isoformat()
finalClinFeatureJSON["type"]="clinicalFeature"
finalClinFeatureJSON["name"]=jsonName
#final clinicalFeature
if finalClinFeatureJSON:
fout = open(".tmp",'w')
fout.close()
for clinF in clinFeatures:
fout = open(".tmptmp",'w')
clinF.store(fout)
fout.close()
os.system("cat .tmptmp >> .tmp")
fin = open(".tmp",'r')
finalClinFeature =ClinicalFeatureNew(fin,finalClinFeatureJSON['name'])
if not finalClinFeature.isValid():
print "final clinFeature file .tmp is invalid"
return 0
fin.close()
#SURVIVAL analysis data
foundE=0
foundT=0
if finalClinFeature:
features= finalClinFeature.getFeatures()
for feature in features:
sameAs = finalClinFeature.getFeatureSameAs(feature)
if sameAs =="_TIME_TO_EVENT":
#check there is only one parameter is set to be _TIME_TO_EVENT
if foundT==1:
print "ERROR there is already _TIME_TO_EVENT"
continue
#check matrix does not have _TIME_TO_EVNET
if sameAs in finalClinMatrix.getCOLs():
print "ERROR there is already _TIME_TO_EVENT in matrix"
continue
#data check need to check these are floats or "" in both clinFeature and clinMatrix
GOOD=1
if finalClinMatrix.isTypeFloat(feature)!= True:
print "ERROR _TIME_TO_EVENT parent feature values are not correct", finalClinMatrix.getColStates(feature)
GOOD=0
if GOOD:
foundT=1
finalClinMatrix.addNewColfromOld(sameAs, feature)
finalClinFeature.setFeatureValueType(sameAs,"float")
if sameAs =="_EVENT":
#check there is only one parameter is set to be _EVENT
if foundE==1:
print "ERROR there is already _EVENT"
continue
#check matrix does not have _EVNET
if sameAs in finalClinMatrix.getCOLs():
print "ERROR there is already _EVENT in matrix"
continue
#data check
GOOD=1
states= finalClinMatrix.getColStates(feature)
"""
for state in states:
if state not in [0,1,"0","1",""]:
print "ERROR _EVENT values are not correct", state
GOOD=0
break
"""
if len(states) not in [2,3]:
GOOD=0
if len(states)==3 and states.count('')!=1:
GOOD=0
if GOOD:
foundE=1
finalClinMatrix.addNewColfromOld(sameAs, feature)
finalClinFeature.setFeatureValueType(sameAs,"category")
#finalClinFeature.setFeatureStates(sameAs,["0","1"])
#finalClinFeature.setFeatureStateOrder(sameAs,["0","1"])
#clinical data push down
roots = sMap.allRoots()
for root in roots:
r = finalClinMatrix.pushToChildren (root,sMap)
if r != True:
print "Fail to push down"
return 0
print "after clinical push down", sampleMap,finalClinMatrix.getROWnum()
# collect all genomic data
keepSamples = getAllGenomicIds(sMap, bookDic)
# removing rows without genomic data from clinical data matrix due to mysql enum limitation
# should remove this step after the display functionality is done better, currently cgb clinical data range in feature control panel shows the full range of clinical data without checking if the specific track/dataset has the full value range.
if onlyGenomicSamples:
print "genomic sample count", len(keepSamples)
success= finalClinMatrix.onlyKeepRows(keepSamples)
if not success:
print "fail to remove extra rows"
else:
print "after keeping sample with genomic data", finalClinMatrix.getROWnum()
#add to the clinical matrix any samples with genomic data but no clinical data
emptyData={}
for col in finalClinMatrix.getCOLs():
emptyData[col]=""
success = finalClinMatrix.addNewRows(keepSamples,emptyData)
if not success:
print "fail to add new roows"
else:
print "after adding all genomic data", finalClinMatrix.getROWnum()
if finalClinMatrix.validate() != True:
print "Fail to validate"
cMatrix = oldCMatrix
return 0
# end of collecting all genomic data
#code to remove blacklist samples and all its descendants
badList= badListSelfAndDescendants (sMap, bookDic)
if badList!=[]:
#remove badList samples
finalClinMatrix.removeRows(badList, True)
print "after remove badList", finalClinMatrix.getROWnum()
#identify empty features
badFeatures= finalClinMatrix.findBadColsNotRemove()
print "emptye features:", badFeatures
#finalBadFeatures=[]
#if finalClinFeature: ########### don't understand this
# for feature in badFeatures:
# #get short label
# shortTitle = finalClinFeature.getShortTitle(feature)
# if not shortTitle:
# print feature,"remove"
# finalBadFeatures.append(feature)
# else:
# print shortTitle,"not remove"
#else:
# finalBadFeatures =badFeatures[:]
#remove bad features
finalBadFeatures= badFeatures
finalClinMatrix.removeCols(finalBadFeatures)
print "remove features", finalBadFeatures
# add _PATIENT col
if finalClinMatrix.addColRoot(sMap) == None:
print "Fail to addColRoot"
return 0
# add _INTEGRATION col
intList=[]
if bookDic.has_key(sampleMap) and bookDic[sampleMap].has_key(":integrationId"):
intName=bookDic[sampleMap][":integrationId"]
fin= open(bookDic[intName]["path"],"r")
intId = IntegrationId(intName,fin)
intList = intId.getList()
finalClinMatrix.addColIntegration(sMap,intList)
# final ClinFeature json
if finalClinFeatureJSON==None:
jsonName= trackName_fix(sampleMapBaseName(sMap)+"_clinicalFeature")
finalClinFeatureJSON= {}
finalClinFeatureJSON["version"]=datetime.date.today().isoformat()
finalClinFeatureJSON["type"]="clinicalFeature"
finalClinFeatureJSON["name"]=jsonName
finalClinFeatureJSON["path"]=""
finalClinFeature = ClinicalFeatureNew (None, finalClinFeatureJSON["name"])
#final clinicalFeature
if finalClinFeature:
finalClinFeature.removeFeatures(finalBadFeatures)
finalClinFeature.cleanState()
finalClinFeature.checkFeatureWithMatrix(finalClinMatrix)
#clinicalFeature fillin ValueType
finalClinFeature.fillInValueTypeWithMatrix(finalClinMatrix)
#clinicalFeature fillin missing features
finalClinFeature.fillInFeaturesWithMatrix(finalClinMatrix)
#clinicalFeature fillin short and long titles
finalClinFeature.fillInTitles()
#clinicalFeature fillin priority visibility
#vis exceptions
VIS_limit=4
if bookDic.has_key(sMap.getName()) and bookDic[sMap.getName()].has_key("VIS"):
VIS_limit= bookDic[sMap.getName()]["VIS"]
finalClinFeature.fillInPriorityVisibility(VIS_limit)
finalClinFeature.setFeatureShortTitle("_PATIENT","_PATIENT_ID")
finalClinFeature.setFeatureLongTitle("_PATIENT","_PATIENT_ID")
finalClinFeature.setFeatureValueType("_PATIENT","category")
finalClinFeature.setFeatureShortTitle("_INTEGRATION","_SAMPLE_ID")
finalClinFeature.setFeatureLongTitle("_INTEGRATION","_SAMPLE_ID")
finalClinFeature.setFeatureValueType("_INTEGRATION","category")
print sampleMap,finalClinMatrix.getROWnum()
return finalClinMatrix,finalClinMatrixJSON, finalClinFeature, finalClinFeatureJSON
def flattenEachSampleMap(sMap, bookDic, onlyGenomicSamples):
sampleMap = sMap.getName()
jsonName = trackName_fix(sampleMapBaseName(sMap) + "_clinicalMatrix")
finalClinMatrix = ClinicalMatrixNew(None, jsonName)
finalClinMatrixJSON = {}
finalClinMatrixJSON["name"] = jsonName
finalClinMatrixJSON["type"] = "clinicalMatrix"
finalClinMatrixJSON["path"] = ""
finalClinMatrixJSON[":sampleMap"] = sampleMap
clinFeatures = []
finalClinFeatureJSON = None
finalClinFeature = None
# add all ids to sMap
sMapChanged = checkIdsAllIn(sMap, bookDic)
#build initial clinical Matrix with sampleMap ids, all with empty data
emptyData = {}
success = finalClinMatrix.addNewRows(sMap.getNodes(), emptyData)
if not success:
print "fail to add all initial ids from sampleMap"
datasets = collectNamesBelongToSampleMap(bookDic, sampleMap)
datasetsOrdered = [] #only the ClinicalMatrix ordered list
for name in datasets:
obj = bookDic[name]
if obj['type'] == "clinicalMatrix":
if obj.has_key('outOfDate') and obj['outOfDate'] in [
"yes", "Yes", "YES"
]:
datasetsOrdered.append(name)
elif not obj.has_key('outOfDate') and not obj.has_key('upToDate'):
datasetsOrdered.insert(0, name)
upToDateSets = {}
for name in datasets:
obj = bookDic[name]
if obj['type'] == "clinicalMatrix":
if obj.has_key('upToDate'):
upToDateSets[obj['upToDate']] = name
keys = upToDateSets.keys()
keys.sort()
for version in keys:
name = upToDateSets[version]
datasetsOrdered.insert(0, name)
for name in datasetsOrdered:
obj = bookDic[name]
if obj['type'] == "clinicalMatrix":
clinFeature = None
#clinFeature obj
if obj.has_key(':clinicalFeature'):
path = bookDic[obj[':clinicalFeature']]['path']
neme = bookDic[obj[':clinicalFeature']]['name']
clinFeature = ClinicalFeatureNew(path, name)
#get matrix obj
path = obj['path']
name = obj['name']
cMatrix = ClinicalMatrixNew(path, name, False, clinFeature)
if finalClinMatrix == None:
finalClinMatrix = cMatrix
if finalClinMatrixJSON == None:
finalClinMatrixJSON = obj
#merge final and cMatrix
if finalClinMatrix != cMatrix:
print "name=", cMatrix.getName()
r = finalClinMatrix.addNewCols(cMatrix, validation=True)
if r != True:
print "Fail to merge"
return False
#add clinFeature
if clinFeature:
clinFeatures.append(clinFeature)
#merge finalClinMatrixJSON with new json
if finalClinMatrixJSON != obj:
jsonName = trackName_fix(
sampleMapBaseName(sMap) + "_clinicalMatrix")
finalClinMatrixJSON = cgDataMergeJSON(finalClinMatrixJSON, obj,
jsonName)
# final ClinFeature json
if clinFeature:
clinFeatureJSON = bookDic[obj[':clinicalFeature']]
if finalClinFeatureJSON == None:
finalClinFeatureJSON = clinFeatureJSON
else:
jsonName = trackName_fix(
sampleMapBaseName(sMap) + "_clinicalFeature")
finalClinFeatureJSON["version"] = datetime.date.today(
).isoformat()
finalClinFeatureJSON["type"] = "clinicalFeature"
finalClinFeatureJSON["name"] = jsonName
#final clinicalFeature
if finalClinFeatureJSON:
fout = open(".tmp", 'w')
fout.close()
for clinF in clinFeatures:
fout = open(".tmptmp", 'w')
clinF.store(fout)
fout.close()
os.system("cat .tmptmp >> .tmp")
fin = open(".tmp", 'r')
finalClinFeature = ClinicalFeatureNew(fin,
finalClinFeatureJSON['name'])
if not finalClinFeature.isValid():
print "final clinFeature file .tmp is invalid"
return 0
fin.close()
#SURVIVAL analysis data
foundE = 0
foundT = 0
if finalClinFeature:
features = finalClinFeature.getFeatures()
for feature in features:
sameAs = finalClinFeature.getFeatureSameAs(feature)
if sameAs == "_TIME_TO_EVENT":
#check there is only one parameter is set to be _TIME_TO_EVENT
if foundT == 1:
print "ERROR there is already _TIME_TO_EVENT"
continue
#check matrix does not have _TIME_TO_EVNET
if sameAs in finalClinMatrix.getCOLs():
print "ERROR there is already _TIME_TO_EVENT in matrix"
continue
#data check need to check these are floats or "" in both clinFeature and clinMatrix
GOOD = 1
if finalClinMatrix.isTypeFloat(feature) != True:
print "ERROR _TIME_TO_EVENT parent feature values are not correct", finalClinMatrix.getColStates(
feature)
GOOD = 0
if GOOD:
foundT = 1
finalClinMatrix.addNewColfromOld(sameAs, feature)
finalClinFeature.setFeatureValueType(sameAs, "float")
if sameAs == "_EVENT":
#check there is only one parameter is set to be _EVENT
if foundE == 1:
print "ERROR there is already _EVENT"
continue
#check matrix does not have _EVNET
if sameAs in finalClinMatrix.getCOLs():
print "ERROR there is already _EVENT in matrix"
continue
#data check
GOOD = 1
states = finalClinMatrix.getColStates(feature)
"""
for state in states:
if state not in [0,1,"0","1",""]:
print "ERROR _EVENT values are not correct", state
GOOD=0
break
"""
if len(states) not in [2, 3]:
GOOD = 0
if len(states) == 3 and states.count('') != 1:
GOOD = 0
if GOOD:
foundE = 1
finalClinMatrix.addNewColfromOld(sameAs, feature)
finalClinFeature.setFeatureValueType(sameAs, "category")
#finalClinFeature.setFeatureStates(sameAs,["0","1"])
#finalClinFeature.setFeatureStateOrder(sameAs,["0","1"])
#clinical data push down
roots = sMap.allRoots()
for root in roots:
r = finalClinMatrix.pushToChildren(root, sMap)
if r != True:
print "Fail to push down"
return 0
print "after clinical push down", sampleMap, finalClinMatrix.getROWnum()
# collect all genomic data
keepSamples = getAllGenomicIds(sMap, bookDic)
# removing rows without genomic data from clinical data matrix due to mysql enum limitation
# should remove this step after the display functionality is done better, currently cgb clinical data range in feature control panel shows the full range of clinical data without checking if the specific track/dataset has the full value range.
if onlyGenomicSamples:
print "genomic sample count", len(keepSamples)
success = finalClinMatrix.onlyKeepRows(keepSamples)
if not success:
print "fail to remove extra rows"
else:
print "after keeping sample with genomic data", finalClinMatrix.getROWnum(
)
#add to the clinical matrix any samples with genomic data but no clinical data
emptyData = {}
for col in finalClinMatrix.getCOLs():
emptyData[col] = ""
success = finalClinMatrix.addNewRows(keepSamples, emptyData)
if not success:
print "fail to add new roows"
else:
print "after adding all genomic data", finalClinMatrix.getROWnum()
if finalClinMatrix.validate() != True:
print "Fail to validate"
cMatrix = oldCMatrix
return 0
# end of collecting all genomic data
#code to remove blacklist samples and all its descendants
badList = badListSelfAndDescendants(sMap, bookDic)
if badList != []:
#remove badList samples
finalClinMatrix.removeRows(badList, True)
print "after remove badList", finalClinMatrix.getROWnum()
#identify empty features
badFeatures = finalClinMatrix.findBadColsNotRemove()
print "emptye features:", badFeatures
#finalBadFeatures=[]
#if finalClinFeature: ########### don't understand this
# for feature in badFeatures:
# #get short label
# shortTitle = finalClinFeature.getShortTitle(feature)
# if not shortTitle:
# print feature,"remove"
# finalBadFeatures.append(feature)
# else:
# print shortTitle,"not remove"
#else:
# finalBadFeatures =badFeatures[:]
#remove bad features
finalBadFeatures = badFeatures
finalClinMatrix.removeCols(finalBadFeatures)
print "remove features", finalBadFeatures
# add _PATIENT col
if finalClinMatrix.addColRoot(sMap) == None:
print "Fail to addColRoot"
return 0
# add _INTEGRATION col
intList = []
if bookDic.has_key(sampleMap) and bookDic[sampleMap].has_key(
":integrationId"):
intName = bookDic[sampleMap][":integrationId"]
fin = open(bookDic[intName]["path"], "r")
intId = IntegrationId(intName, fin)
intList = intId.getList()
finalClinMatrix.addColIntegration(sMap, intList)
# final ClinFeature json
if finalClinFeatureJSON == None:
jsonName = trackName_fix(sampleMapBaseName(sMap) + "_clinicalFeature")
finalClinFeatureJSON = {}
finalClinFeatureJSON["version"] = datetime.date.today().isoformat()
finalClinFeatureJSON["type"] = "clinicalFeature"
finalClinFeatureJSON["name"] = jsonName
finalClinFeatureJSON["path"] = ""
finalClinFeature = ClinicalFeatureNew(None,
finalClinFeatureJSON["name"])
#final clinicalFeature
if finalClinFeature:
finalClinFeature.removeFeatures(finalBadFeatures)
finalClinFeature.cleanState()
finalClinFeature.checkFeatureWithMatrix(finalClinMatrix)
#clinicalFeature fillin ValueType
finalClinFeature.fillInValueTypeWithMatrix(finalClinMatrix)
#clinicalFeature fillin missing features
finalClinFeature.fillInFeaturesWithMatrix(finalClinMatrix)
#clinicalFeature fillin short and long titles
finalClinFeature.fillInTitles()
#clinicalFeature fillin priority visibility
#vis exceptions
VIS_limit = 4
if bookDic.has_key(
sMap.getName()) and bookDic[sMap.getName()].has_key("VIS"):
VIS_limit = bookDic[sMap.getName()]["VIS"]
finalClinFeature.fillInPriorityVisibility(VIS_limit)
finalClinFeature.setFeatureShortTitle("_PATIENT", "_PATIENT_ID")
finalClinFeature.setFeatureLongTitle("_PATIENT", "_PATIENT_ID")
finalClinFeature.setFeatureValueType("_PATIENT", "category")
finalClinFeature.setFeatureShortTitle("_INTEGRATION", "_SAMPLE_ID")
finalClinFeature.setFeatureLongTitle("_INTEGRATION", "_SAMPLE_ID")
finalClinFeature.setFeatureValueType("_INTEGRATION", "category")
print sampleMap, finalClinMatrix.getROWnum()
return finalClinMatrix, finalClinMatrixJSON, finalClinFeature, finalClinFeatureJSON
