Citation

Material Information

Title:
Cognition and connectomes in nondementia idiopathic Parkinson’s disease
Series Title:
Colon-Perez, L. M., Tanner, J. J., Couret, M., Goicochea, S., Mareci, T. H., & Price, C. C. (2017). Cognition and connectome in nondementia idiopathic Parkinson’s disease. Network Neuroscience. 2(1) 106–124. https://doi.org/10.1162/netn_a_00027
Creator:
Colon-Perez, Luis
Publisher:
Network Neuroscience
Publication Date:
Language:
English
Physical Description:
Journal Article

Notes

Abstract:
In this study, we investigate the organization of the structural connectome in cognitively well participants with Parkinson’s disease (PD-Well; n = 31) and a subgroup of participants with Parkinson’s disease who have amnestic disturbances (PD-MI; n = 9). We explore correlations between connectome topology and vulnerable cognitive domains in Parkinson’s disease relative to non-Parkinson’s disease peers (control, n = 40). Diffusion-weighted MRI data and deterministic tractography were used to generate connectomes. Connectome topological indices under study included weighted indices of node strength, path length, clustering coefficient, and small-worldness. Relative to controls, node strength was reduced 4.99% for PD-Well (p = 0.041) and 13.2% for PD-MI (p = 0.004). We found bilateral differences in the node strength between PD-MI and controls for inferior parietal, caudal middle frontal, posterior cingulate, precentral, and rostral middle frontal. Correlations between connectome and cognitive domains of interest showed that topological indices of global connectivity negatively associated with working memory and displayed more and larger negative correlations with neuropsychological indices of memory in PD-MI than in PD-Well and controls. These findings suggest that indices of network connectivity are reduced in PD-MI relative to PD-Well and control participants.
Acquisition:
Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Luis Colon-Perez.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
Copyright Creator/Rights holder. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

RESEARCH Cognitionandconnectomesinnondementia idiopathicParkinsonsdisease LuisM.Colon-Perez 1 ,JaredJ.Tanner 2 ,MichelleCouret 3 ,ShelbyGoicochea 4 ThomasH.Mareci 5 ,andCatherineC.Price 2 1 DepartmentofPsychiatry,UniversityofFlorida,Gainesville,FL,USA 2 DepartmentofClinicalandHealthPsychology, UniversityofFlorida,Gainesville,FL,USA 3 DepartmentofMedicine,Columbi aUniversity,NewYork,NY,USA 4 DepartmentofMedicine,University ofFlorida,Gainesville,FL,USA 5 DepartmentofBiochemistryandMolecularBiolo gy,UniversityofFlorida,Gainesville,FL,USA Keywords: Connectome,Connectivity,Tractography,Structuralnetworks,Parkinsonsdisease, Cognitivedecline ABSTRACT Inthisstudy,weinvestigatetheorganizationofthestructuralconnectomeincognitively wellparticipantswithParkinsonsdisease(PD-Well; n = 31 )andasubgroupof participantswithParkinsonsdiseasewhohaveamnesticdisturbances(PD-MI; n = 9 ). Weexplorecorrelationsbetweenconnectometopologyandvulnerablecognitive domainsinParkinsonsdiseaserelativetonon-Parkinsonsdiseasepeers(control, n = 40 ). Diffusion-weightedMRIdataanddeterministictractographywereusedtogenerate connectomes.Connectometopologicalindicesunderstudyincludedweightedindicesof nodestrength,pathlength,clusteringcoef“cient,andsmall-worldness.Relativetocontrols, nodestrengthwasreduced4.99%forPD-Well( p = 0.041 )and13.2%forPD-MI( p = 0.004 ). WefoundbilateraldifferencesinthenodestrengthbetweenPD-MIandcontrolsforinferior parietal,caudalmiddlefrontal,posteriorcingulate,precentral,androstralmiddlefrontal. Correlationsbetweenconnectomeandcognitivedomainsofinterestshowedthattopological indicesofglobalconnectivitynegativelyassociatedwithworkingmemoryanddisplayed moreandlargernegativecorrelationswithneuropsychologicalindicesofmemoryinPD-MI thaninPD-Wellandcontrols.These“ndingss uggestthatindicesofnetworkconnectivityare reducedinPD-MIrelativetoPD-Wellandcontrolparticipants. AUTHORSUMMARY Parkinsonsdisease(PD)patientswithamnesticmildcognitiveimpairment(e.g.,primary processing-speedimpairmentsorprimarymemoryimpairments)areatgreaterriskof developingdementia.RecentevidencesuggeststhatpatientswithPDandmildcognitive impairmentpresentanalteredconnectomeconnectivity.Inthiswork,wefurtherexplorethe structuralconnectomeofPDpatientstoprovidecluestoidentifypossiblesensitivemarkers ofdiseaseprogression,andcognitiveimpairment,insusceptiblePDpatients.Weemployeda weightednetworkframeworkthatyieldsmorestabletopologicalresultsthanthebinary networkframeworkandisrobustdespitegraphdensitydifferences,henceitdoesnotrequire thresholdingtoanalyzetheconnectomes.AsSupplementaryInformation( Colon-Perezetal. 2017 ),weincludedatabasessharingtheresultsofthenetworkdata. anopenaccess journal Citation:Colon-Perez,L.M.,Tanner, J.J.,Couret,M.,Goicochea,S.,Mareci, T.H.,&Price,C.C.(2017).Cognition andconnectomeinnondementia idiopathicParkinson’sdisease. NetworkNeuroscience 2 (1)106–124. https://doi.org/10.1162/netn_a_00027 DOI: https://doi.org/10.1162/netn a 00027 SupportingInformation: https://doi.org/10.1162/netn a 00027 Received:18February2017 Accepted:18September2017 CompetingInterests:Theauthorshave declaredthatnocompetinginterests exist. CorrespondingAuthor: LuisM.Colon-Perez lcolon@u.edu Copyright: 2017 MassachusettsInstituteofTechnology PublishedunderaCreativeCommons Attribution4.0International (CCBY4.0)license TheMITPress

PAGE 2

ConnectomesinParkinson’sdisease INTRODUCTIONParkinsonsdisease(PD)isaneurodegenerativedisordercharacterizedbymotordisruption (i.e.,tremors,unstableposture,bradykinesia)andcognitivesymptoms( Chaudhuri&Schapira 2009 ).AlthoughJamesParkinsoninitiallyproposedthatPDdidnotprogresstothecerebrum ( Parkinson 2002 ),thereisnowsubstantialevidencethatPDcancompromisehighercortical cognitiveprocesses.TheprogressionmaynotbethesameforallindividualswithPD,however; someindividualsremainwithprimaryprocessing-speedimpairmentswhileothersmaypresent withprimarymemoryimpairmentsandareatgreaterriskfordementia( Hendersonetal. 2016 ; Janvin,Larsen,Aarsland,&Hugdahl 2006 ).Neuroimaginginvestigationsshowregionalfractionalanisotropy(FA)andvolumetricgrayandwhitematter(WM)differencesinPDand non-PDpeers( Priceetal. 2016 ; Tanneretal. 2015 ).PDmaybeanetwork-leveldisease ( Belluccietal. 2016 ; Catani&Ffytche 2005 ; Gratwicke,Jahanshahi,&Foltynie 2015 ),and thenetworkheterogeneitypossiblyexplainswhetherthesubjectshavememoryimpairment. Novelapplicationswithstructuralconnectomesholdpromiseforimprovingourunder-Structuralconnectome: Brainconnectivitygraphsobtained fromdiffusionMRIandtractography.standingofnetworkheterogeneitywithinPD.Connectomestudiesrepresentthebrainasaset ofnodes(brainareas)andedges(connectingwhitematterbetweenbrainareas)thatquantify themacroscopictopologicalorganizationofthebrainnetwork.Thetopologicalfeaturesof thehumanconnectomeallowustodescribethecomplexinterconnectednessofthehuman braininvivo.Connectomestudiesquantitativelydescribethearrangementofconnectionsin thebrain( Sporns 2011b )andofferanovelapproachtoexplorethebraininhealthyandneuropathologicalparticipants( Hagmannetal. 2010 ; Sporns 2011a ).Theyhavebeenusedto quantifytheorganizationofconnectedwhitematterinneurologicaldisorderssuchasHuntingtonsdisease( Odishetal. 2015 ),epilepsy( Taylor,Han,Schoene-Bake,Weber,&Kaiser 2015 ),andAlzheimersdisease( Daianuetal. 2015 ).Also,connectometopologyhasbeen suggestedasasensitivebiomarkerforearlystagesofpsychoticillnessandtheeventualdevelopmentofpsychosis( Drakesmithetal. 2015 ).InindividualswithPDwhohaveafreezing gait,maladaptivebrainrestructuringhasbeenshownthroughtheconnectivitybetweenlocomotorhubs,particularlyinthesupplementarymotorareaandmesencephaliclocomotor regions( Flingetal. 2014 ).ArecentstudyshowedthattheconnectomeinpatientswithPD withmildcognitiveimpairment(MCI)isaltered( Galantuccietal. 2016 ).Hence,furtherstructuralconnectomestudiesofPDmayyieldsensitivemarkersofdiseaseprogression,cognitive impairment,andsusceptibilitytoPD. Inthiswork,weperformedglobal(averagevaluesfortheentirebrainnetworkfor eachparticipant)andlocal(averagevaluesforeveryindividualnodeforeachparticipant)connectomeanalyses.Giventheelevatedriskofdevelopingdementiaassociatedwith amnesticmildcognitiveimpairmentinPD( Hendersonetal. 2016 ),inthisstudyweexaminedstructuralconnectomedifferencesinpeoplewithPDwhoarecognitivelywell relativetoindividualswithPDwhomeetcriteriaforamnesticMCI.Wealsoexplored thecorrelationsbetweentopologicalconnectomeindicesandthemostcommoncognitivevulnerabilitiesofPD(i.e.,processingspeed,workingmemory,andepisodicmemory; Zgaljardic,Borod,Foldi,&Mattis 2003 ).WeuseddiffusionMRI( Basser&Jones 2002 )and tractography( Basser,Pajevic,Pierpaoli,Duda,&Aldroubi 2000 )togenerateconnectomesTractography: Three-dimensionalrepresentationof thebrainswhitemattertractsderived fromdiffusionMRIdata.comprising82brainregions(68corticaland14subcortical).Withthis,wedeterminedthe organizationofstructuralconnectivityofourthreeparticipantgroups(PD,PD-MCI,non-PD peers)andcorrelationsbetweenconnectomeindicesandworkingmemory,processingspeed, andverbalmemory.NetworkNeuroscience 107

PAGE 3

ConnectomesinParkinson’sdisease METHODSParticipantsThisstudywasapprovedbytheUniversityofFloridaHealthCenterInstitutionalReviewBoard (Protocol#472-2007).Writtenconsentwasobtainedfromallparticipants,andallresearch followedtheDeclarationofHelsinki. ProviderswithintheUFCenterforMovementDisordersandNeurorestorationreferred nondementedindividualswithidiopathicPDtothestudy.Structuredtelephonescreening wasperformedtoverifyinclusion/exclusioncriteria.Potentialparticipantswerescreened inpersonwiththeDementiaRatingScale-2(DRS-2)( Matteauetal. 2011 );atotalDRS-2DementiaRatingScale-2: Ascreeningmeasureusedtoassessa patientsoveralllevelofcognitive functioning.score>130wasrequiredforparticipation.Onlynondementedindividualswhowereable toconsenttoparticipatewereincludedinthestudy.Inclusioncriteria:right-handed ( Briggs&Nebes 1975 ),DRS-2rawscore>130,”uentEnglish,diagnosisofPDbyamovementdisorderneurologist,UKParkinsonsDiseaseSocietyBrainBankClinicalDiagnosticCriteria( Hughes,Ben-Shlomo,Daniel,&Lees 2001 ),andHoehnandYahrscale( Hoehn&Yahr 1967 )rangingfrom1to3.Exclusioncriteria: diseaseslikelytoconf oundcognition(e.g., cerebrovascularaccidentinthelastsixmonths ),deepbrainstimulati on,secondary/atypical Parkinsonism,andmajorpsychiatricdisorder.DepressionandapathywerenotexclusioncriteriabecauseoftheirhighprevalenceinPD. The“nalsampleincluded40peoplewithidiopathicPDand40non-PDpeers.Diffusion andgraymatterstructuraldatafromsomeoftheseparticipantshavealsobeenseeninrecent publications( Crowleyetal. 2017 ; Priceetal. 2016 ; Schwabetal. 2015 ; Tanner,Levy,etal. 2017 ; Tanneretal. 2015 ; Tanner,McFarland,etal. 2017 ).De“ningPDSubgroupsPD-MemoryImpaired(PD-MI):Fromthemeasuresdescribedbelow,thosewithPDwhohada memorycompositescoreŠ1.5 (relativetonon-PDpeers)wereclassi“edasPD-MI( n=9 ). AllotherswithPDwereclassi“edasPDwithoutmemoryimpairment(PD-Well; n=31 ). ThosewithandwithoutPD-MIarediscussedelsewhere( Tanneretal. 2015 ).CognitiveMeasuresWhileonmedication,participantscompletedcognitivetesting,neuroimaging,andtheUni“edParkinsonsDiseaseRatingScale(UPDRS)toassessoptimalperformanceandrepresentUni“edParkinsonsDiseaseRating Scale: Themostcommonlyusedmeasureof Parkinsonsdiseaseclinicalsymptom severity.normalfunctioning.Allparticipantsalsocompletedtestsongeneralcognitionandmood, PDsymptomsandseverity,comorbidity( Charlson,Pompei,Ales,&MacKenzie 1987 ),and aneuropsychologicalprotocol.Medicationswererevertedtoacommonmetric(levodopa equivalencydose,LED; Tomlinsonetal. 2010 ).Blindedratersscoredthedatatwice.TheUPDRSIIIisameasureofPDmotorsymptomseverityandwasusedasacorrelatewithnetwork indices. Primarycognitivedomainsofinterest:PDisbraindisorderofthefrontal-subcorticalareas knowntoalterthefrontal-striatalcognitivefunctionsofprocessingspeedandworkingmemory. Thesedomainswereassessedusingcompositesofstandardizedneuropsychologicalmeasures: Processingspeed:basedonstandardizedscoresfromtheTrailMakingTest,PartA(total time; Heaton&PsychologicalAssessmentResourcesInc. 2004 ),WAIS-IIIDigitSymbol (totalcorrect; Wechsler 1991 ),andStroopColorWordTest…WordReadingcondition (totalcorrect; Golden&Freshwater 2002 ).NetworkNeuroscience 108

PAGE 4

ConnectomesinParkinson’sdisease Workingmemory:createdfromtheDigitSpanBackward(totalspan; Wechsler 1991 ), SpatialSpanBackward(totalscore; Wechsler 1991 ),andLetterNumberSequencing (totalcorrect; Wechsler 1991 ). Inadditiontofrontal-striatald e“cits,weexaminedconnectomei ndicesrelativetodeclarative memoryabilities,asthisisanessentialdomainofprodromaldementia.Prodromaldementia: Theearlieststageof neurodegenerativediseasewhen thereisadeclineinmemoryor cognition,butfunctional independenceremainsintact. Verbalmemory:createdfromselectedindexscoresfromthe12-wordversionofthe Philadelphia(repeatable)VerbalLearningTest(P(r)VLT( Priceetal. 2009 )andWechsler MemoryScale3rdRevision(WMS-III)LogicalMemory(LM)( Wechsler 1991 ).MRIAcquisitionandProcessingDatawereacquiredwithaSiemens3TVeriousinganeight-channelheadcoil.TwoT1weightedscanswereusedfornodesegmentationwithscanparametersof176contiguous slices, 1 mm3isotropicvoxels,andTR/TE=2,500/3.77ms.Single-shotechoplanarimagingdiffusion-weightedimageswereacquiredfor tractographywithgradientsappliedalong6 ( b=100 s/mm2)and64directions( b=1,000 s/mm2).Diffusionimagingparameterswereset at73contiguousaxialsliceswith 8 mm3isotropicvoxelsandTR/TE=17,300/81ms.Node segmentationwascomplete dwithFreeSurfer5.3anddatawerequalitychecked.Thequalitycheckforthediffusiondataincludedvisualinspectionforartifacts(e.g.,signallossinthe anteriorandmiddleregions,Venetianblinding,checkerboarding).Nosigni“cantartifacts wereobserved.Theprocesswasalsorepeatedaftereddycurrentcorrection(eddy_correct). Participantheadmotionduringdiffusionsequenceswasquanti“edwithfourmeasuresusing TRACULA( Yendiki,Koldewyn,Kakunoori,Kanwisher,&Fischl 2013 ).Between-groupregistrationandintensity-basedmetricsdemonstratednosigni“cantgroupdifferencesindiffusion sequencemotion(Registration:averagetranslation: t=0.98 p=0.33 ;averagerotation: 2=1.25 p=0.26 ;Intensity:Percentagebadslices 2=0.26 p=0.61 ;Averagedropout score 2=0.26 p=0.61 ),suggestingthatdatawereappropriateforgroupcomparisons. Diffusiondatawerepreprocessedusingin-housesoftwarewritteninIDL(Harris GeospatialSolutions,Bloom“eld,CO).EddycurrentcorrectionwasperformedusingFSL ( Jenkinson,Beckmann,Behrens,Woolrich,&Smith 2012 ).Diffusiontensorimagingmetrics(fractionalanisotropy,FA,andmean diffusivity,MD)werecalculatedusingFSL. Fortractography,“berorientationpro“leswereestimatedbasedonthecalculationof diffusiondisplacementpr obabilitywithamixtureofth eWishartmethodoutlinedby Jian,Vemuri,Ozarslan,Carney,andMareci ( 2007 ).Diffusionimageswereinterpolated ( Meijering,Zuiderveld,&Viergever 1999 )to 1 mm3isotropicvoxelsusingcubicconvolution andwhole-braindeterministic“bertrackinginitiatedusing125uniformlydistributedstreamlinepointspervoxel.NetworkPreparationandAnalysisThenetworkedgeswereweightedasdescribedbyColon-Perezetal.( Colon-Perez,Spindler, etal.,2015 ).Theedgeweight, w(e),de“nesconnectinganytwonodesisde“nedasEdgeweight: Graphtheoreticalrepresentationof thestrengthofthepairwise connectionsbetweenthenodesina graph.w(eij)= Vvoxel Pvoxel 2 Ai+AjPvoxelp=1 Mm=1R(fp m) l(fp m), (1)NetworkNeuroscience 109

PAGE 5

ConnectomesinParkinson’sdisease where R(fp m)= 1, fp mR 0, fp m/R (2) VvoxelistheMRvoxelvolume, Pvoxelisthenumberofstreamlineseedpointspervoxel, A isthesurfaceareaofeachnode, M isthenumberofvoxelsmakinguptheedge, fp mis thestreamlineoriginatingfromseedpoint p invoxel m l(fp m)isthelengthof fp m,Ris thesetofstreamlinesthatoriginatefromthevoxelsmakingupthespaceoccupiedbythe WMpathconnectingnodes niand nj,and R(fp m)isthecharacteristicfunctionthaten-Characteristicfunction: Inprobabilityandstatistics,a functionthatpreciselyde“nesthe membersofaset.Ittakesavalueof oneforamemberofthesetandzero foranonmember.suresthestreamlinesconnectingnodes niand njoriginatefromthespace(i.e.,voxels)the streamlinestraversebetweennodes niand nj.Theedgeweight(Equation1)eliminatesthe biaseffectsofthelengthofthestreamlines,theseedingparadigm(i.e., Pvoxel),imageresolution(i.e., Vvoxel),andtractography-speci“cexperimentalfactorsfromthecalculation;formore detailswereferthereaderto Colon-Perez,Spindler,etal. ( 2015 ).Thecharacteristicfunction ( R(fp m),Equation2)eliminatesthosestreamlinesthatoriginatewithinthenodesandvoxels thatdonotrepresenttheWMpathconnectingthenodes.Also,thisedgeweightquanti“es thewhitematterstrengthbetweenanytwonodesinadimensionlessandscale-invariantmanner( Colon-Perez,Spindler,etal. 2015 ).Theemployededgeweightusesthestrictcriterionof R(fp m)todeterminethesetstreamlinesusedtoquantifythestrengthofconnectivitybetween twonodes.Giventhehighleveloffalsepositivesintractography,thisedgeweightservesasa layerofstrictcontroltoquantifythestrengthofconnectivitybetweennodes. Thenetworkswereanalyzedinaweightedframework,asdescribedin Colon-Perez,Couret, Triplett,Price,andMareci(2016) .Thisapproachissimilartothebinaryframeworkusedby WattsandStrogatz ( 1998 ),butwithanadditionaldegreeoffreedomfromtheedgeweighting(Equation1).Inourpreviouswork,weshowedthatthisweightedframeworkyields topologicallyrelevantfeatureswithouttheneedforthresholds,thereforenothresholdwas appliedtogeneratethebrainconnectomesinthiswork( Colon-Perezetal. 2016 ).Weak edgesinanetworkarethoughttoprovideacohesivestrengthtonetworks( Granovetter 1973 ). Theabilitytoobtainstableconnectomeresultsacrossthresholds,shownin Colon-Perezetal. ( 2016 ),allowsustomaintaintheseweakedgesinouranalysis.Thesenetworkswerestudied withthefollowingindices:(a) graphdensity ( Boccaletti,Latora,Moreno,Chavez,&Hwang 2006 ),whichisabinarymetricthatquanti“esthefractionofedgesinagraph(only nonweighted indexinthisstudy);(b) nodestrength ( Newman 2001 ),whichisaweightedtopologicalindexoftherelativeconnectivitystrengthofthenodeswiththerestofthenetwork;(c) clusteringcoef“cient ( Zhang&Horvath 2005 ),whichisaweightedmetricthatquanti“esthe strengthofconnectivitybetweentheneighborsofanode;(d) pathlength ( Colon-Perezetal. 2016 ),whichisaweightedmetricthatquanti“esthestrengthoftheshortestpathbetweentwo nodes;and(d) small-worldness ( Humphries&Gurney 2008 ),whichisaweightedmetricthat estimatesthelikelihoodthatnetworksdisplaysimilarpathlengthsandhigherclusteringtoa networkconnectedbyrandomlyassignededges,asdescribedby ErdsandRnyi(1959) .For acompletedescriptionoftheweightednetworkanalysis,referto Colon-Perezetal. ( 2016 ). Theresultsofthesenetworkindicesinthismanuscriptwillbereferredtoasglobalforeach participantwhentheresultsareaveragedintoasinglevaluefortheentirebrainnetwork (yieldsonevalueperparticipant).Thelocalresultsforeachparticipantrefertotheaverage valuespernode(yields82valuesperparticipant).PreviousstudiesdescribedglobaldifferencesinnodestrengthandpathlengthinpatientswithPD( Galantuccietal. 2016 );in thiswork,weusetheseindicestoidentifylocalchangesinnetworkconnectivityinaddition tocorroboratepreviousglobalchangesdescribedby Galantuccietal. ( 2016 ).Theclusteringcoef“cientisreducedgloballyinpatientswithPD( Luoetal. 2015 );inthisstudy,weNetworkNeuroscience 110

PAGE 6

ConnectomesinParkinson’sdisease identifythelocalchangesinclusteringcoef“cientandidentifythosenodesresponsibleforthe reduction.SeveralPDstudieshavecon“rmedthesmall-worldtopologyincontrolsandpatientswithPD;inthisstudy,weassesswhetherPD-WellorPD-MIdisplaysadifferencefrom controls.StatisticsandCorrelationsCognitivecompositesandnetworkvariablesweretestedforstatisticalsigni“cancewithanon-Cognitivecomposites: Averagesofstandardizedscoresfrom multipleneurocognitivemeasures theoreticallymeasuringthesame cognitivedomain.parametricMann-WhitneytestusingR(version3.1.3)( Team 2015 ).Allnode-speci“cnetwork resultswerecontrolledbythegraphdensitytocontrolthein”uenceofglobalconnectivity featuresinlocalindices(i.e.,node-speci“c; vanWijk,Stam,&Daffertshofer 2010 ).Tocorrectformultiplecomparisons,weusedthefdrtoolŽpackage( Strimmer 2008 )andthefalse nondiscoveryrate,whichestimatestheproportionofnondiscoveryrejectionsortypeIIerrors ( Genovese&Wasserman 2002 ).Thistoolworksby“rst“ndingasuitablecutoffpointusing anapproximatenullmodel,whichis“tted;subsequently,acutoffpointissoughtwithafalse nondiscoveryrateassmallaspossible.Scaleparametersofthenullmodelandproportionof nullvaluesarethenestimatedfromthedata.Thecorresponding p valuesarecomputed,anda modi“edGrenanderalgorithm(The Grenander 1956 ,1956,Rimplementationinfdrtoolcan befoundinhttps://cran.r-proj ect.org/web/packages/fd rtool/fdrtool.pdf)isusedto“ndtheoveralldensityanddistributionfunction.Finally,adjusted p valuesaredeterminedandreported. Correlationswerecalculatedbetweennetworkindices(e.g.,nodestrength,pathlength,clustering,small-worldness)andneuropsychologicalcomposites(e.g.,workingmemory,memory speed,andverbalmemory).Sinceourgoalistoidentifythenetworkindicesthatcorrelatewith thedifferentneurocognitivedomainsunderstudy,nocorrelationswereassessedbetweenthe variousnetworkindices.ThecorrelationswereestimatedusingaSpearmanspartialcorrelationmethodcontrollingforeducationlevelus ingtheppcorŽpackageinR.Withsigni“cance inourcorrelativeanalysissetat =0.05 andgivenoursamplesizeof N=80 ,wedecided toreducepotentialfalsepositivesbyconsideringonlycorrelationslargerthan0.50( Cohen 1992 ).RESULTSDemographicsControlandPDgroupswerenotstatisticallydifferentintheiraverageage( Table1 ).ThePDWellgroupdidnotshowasigni“cantdifferencefromcontrolsineducation,workingmemory, ormemorycomposites( Table1 ). ThePD-Wellparticipantsworking-memoryscoresdidnotshowsigni“cantde“cits,whereas PD-MIwas0.88standarddeviationslowerinworking-memoryscore.ThePD-MIgroupwas 2.5yearslesseducatedthancontrols( Table1 )andPD-Well( p=0.03 p valueobtainedbetweenPD-MIandPW-Well).Relativetocontrols,PD-WellandPD-MIdisplayedareduction intheirprocessing-speedscores( Table1 ). ThecombinedPDgroup( n=40 )hadlowermemorycompositescoresthancontrols ( p<0.01 ).PD-Wellhadlowermemorycompositescoresthancontrols,butthiswasnotstatisticallysigni“cant( p=0.082 ; Table1 ).PD-MIhadlowermemorycompositescoresthanboth cohorts:controls( p<0.001 )andPD-Well( p<0.001 ; Table1 ).Groupdifferencesalso remainedaftercontrollingforprocessingspeedandworkingmemory( p<0.001 ).NetworkNeuroscience 111

PAGE 7

ConnectomesinParkinson’sdisease Table1. Demographicsandcognitiveindices Control( n n n=40 )PD-Well( n n n=31 ) p p p PD-MI( n n n=9 ) p p p Mean Mean Mean Age(yrs)68.184.6467.35.020.3869.46.770.60 Duration(yrs)NANA7.875.60NA6.223.07NA Education(yrs)16.752.3516.82.910.8614.32.748.23E-3* UPDRSIII2.753.3618.211.61.34E-10*15.77.254.52E-6* Workingmem1.150.550.890.770.1480.270.576.70E-5* Procspeed0.160.47Š0.42 0.563.23E-5*Š0.68 0.763.90E-3* Memcomp0.001.00Š0.32 0.700.082Š1.76 0.203.54E-5* Statisticalsigni“cancewastestedbetweenPDsubgroups(i.e.,PD-WellandPD-MI)andcontrols. Yrs=years,Duration=diseaseduration,UPDRS=Uni“edParkinsonsDiseaseRatingScale, Workingmem=working-memoryscore,Procspeed=processing-speedcompositescore,Mem comp=memorycompositescore,and =standarddeviation.StatisticsperformedwithMannWhitneystatisticsbetweenPDsandcontrols.*correspondtostatisticallysigni“cantdifferences.GlobalNetworkMeasures ResultsanddifferencesTheconnectomesin Figure1A showthetop1%ofthestrongestconnectionsandtheconnectivityalterationsinthePD-MIconnectomesrelativetocontrols.The nodesizeandcolorrepresentthenodestrength,andtheedgethicknessrepresentstheedge weightvalue.Inparticular,connectionsbetweennodesinthetemporallobearesmallerin size(representingareductioninedgeweight).Thegraphdensity( Figure1B )inallgroups wasapproximately40%ofallpossibleedges,withnosigni“cantdifferencesbetweencontrolsandPD-Well( p=0.249 )orbetweencontrolsandPD-MI( p=0.624 ).Toreducethe biasofgraphdensityonnetworkindices,thesewillbecontrolledbythegraphdensityforthe restofthemanuscript( vanWijketal. 2010 ).A4.99%reductioninmeannodestrengthis observedovertheentirenetworkbetweencontrolsandPD-Well( p=0.041 ),whereasmean nodestrengthinthePD-MIgroupwasreducedrelativetocontrolsby13.23%( p=0.004 ; Figure1C ).Themeanpathlengthsfortheentirenetworkwerereducedby1.97%between controlsandPD-Well( p=0.058 ),whereasthemeanpathlengthsinthePD-MIgroupdisplayedasigni“cantdecreaseof11.7%( p=0.014 Figure2B ).Themeanclusteringcoef“cient fortheentirenetworkdidnotdiffersigni“cantlybetweencontrolsandPD-Well( p=0.162 )or PD-MI( p=0.320 ),withmeanvaluesapproximately0.55( Figure2A ).Thesmall-worldnessindexdidnotdiffersigni“cantlybetweencontrolsandPD-Well( p=0.188 )orPD-MI( p=0.131 ), withmeanvaluesapproximately8.00( Figure2C ).Also,wedeterminednodestrength,clusteringcoef“cient,andpathlengthwithamoretraditionalweightingschemeofFAastheedge weight.Wedidnot“ndanydifferencesbetweencontrolsandPD-WellorcontrolsandPD-MI usingFAastheedgeweight(all p>0.1 ;thedatabaseisavailableasanonlinesupplement andSupplementaryFigure3;see Colon-Perezetal. 2017 ).CorrelationsForthecontrols,globalnetworkindicesdidnotshowanycorrelationswiththeGlobalnetworkindices: Averagevalueofanetworkindex acrossallnodesforasinglesubject.primaryneuropsychologicalcompositesofinterest( Table2 ).Thememorycompositescoredid notcorrelatewithanynetworkindexforanyofthegroups.Theworking-memorycomposite scoreshowednegativecorrelationswithnodestre ngthforPD-Wellpar ticipantsandcontrol. PathlengthalsoshowedanegativecorrelationwithworkingmemorybetweenPD-Welland controls.ThePD-MIgroupshowednegativecorrelationsbetweenallnetworkindicesandthe processing-speedcomposite(exceptsmall-worldness);workingmemorynegativelycorrelated withpathlength.ForPD-MI,therewerepositivecorrelationsbetweentheUni“edParkinsonsNetworkNeuroscience 112

PAGE 8

ConnectomesinParkinson’sdisease Figure1. Thestructuralconnectivityofcontrols,PD-Well,andPD-MIparticipants.(A)Average connectomepergroup;thenodesizerepresentsthestrengthoftheconnectionswiththerestof thenetwork,andtheedgewidthrepresentstherelativestrengthofconnectionsbetweenpairsof nodes.(B)Box-plotdistributionsofgraphdensityvalues(numberofconnectionsinconnectome). Nosigni“cantdifferencesinthenumberofedgeswereobservedbetweengroups.(C)Box-plot distributionsofaverageglobalnodestrength.Themeannodestrengthgroupcomparisonbetween ControlandPD-MIwassigni“cantlyreduced( p=0.002 ).Theconnectomeimagesinthis“gure werepreparedusingBrainNet( Xia,Wang,&He 2013 ). Figure2. Boxplotsofglobalnetworkindicesforallgroups.(A)Meanclustering=mean globalclusteringcoef“cientperbrainaveragedacrosssubjectswithineachgr oup.(B)Meanpath length=meanglobalpathlengthperbrainaveragedacrosssubjectswithineachgr oup. (C)Small-worldness=small-worldnessindexperbrainaveragedacrosssubjectswithineachgr oup. *=statisticallysigni“cantdifference( p<0.05 ),=outliers.StatisticsperformedwithanonparametricMann-Whitneytest.NetworkNeuroscience 113

PAGE 9

ConnectomesinParkinson’sdisease Table2. Correlations(Fishersz-score)betweenglobalnetworkindicesandneuropsychological composites Networkindex Controls( n n n=40 ) UPDRSWorkingmemProcspeedMemcomp NodestrengthNA0.2250.0210.230 ClusteringNA0.0790.1490.195 PathlengthNA0.1000.0510.082 Small-worldnessNAŠ0.144Š0.076Š0.148 Networkindex PD-Well( n n n=31 ) UPDRSWorkingmemProcspeedMemcomp Nodestrength0.008Š0.5740.169Š0.179 ClusteringŠ0.287 0.0180.2510.076 Pathlength0.070Š0.5940.193Š0.229 Small-worldnessŠ0.172 0.1150.230Š0.049 Networkindex PD-MI( n n n=9 ) UPDRSWorkingmemProcspeedMemcomp Nodestrength 0.824#Š0.314Š0.594#0.116 ClusteringŠ0.071Š0.328Š0.685#0.356 Pathlength 0.955#Š0.826Š0.724#0.312 Small-worldnessŠ0.084Š0.109Š0.450 0.079 UPDRS=Uni“edParkinsonsDiseaseRatingScale,Workingmem=working-memoryscore, Procspeed=processing-speedcompositescore,Memcomp=memorycompositescore,and clustering=clusteringcoef“cientscore.PartialcorrelationcontrolledforageusingSpearmans methodtransformedtoFishersz-scoresusingthepsychŽpackageinR.Blackbinscorres pond to|z| >0.55 andcorrelatedresults.*correspondstostatisticallysigni“cantdifferencetocontrols;#correspondstostatisticallysigni“cantdifferencetoPD-Well,calculatedfrom r valueswith http://vassarstats.net/rdiff.html.DiseaseRatingScale(UPDRS)PartIII(motorte st)andnodestrength,aswellaspathlength ( Table2 ;scatterplotsareshowninSupplementaryFigure2;see Colon-Perezetal. 2017 ).LocalNetworkMeasures GroupdifferencesNodestrengthandpathlengthboxplotsforeachgroupareshownin Figures3 and 4 .Therewerelocalnetwork(i.e.,node-speci“c)differencesbetweencontrolsLocalnetworkindices: Valueofanetworkindexforevery nodeandasinglesubject.andPD-MI(nodifferencebetweencontrolsandPD-Well;forspeci“cnoderesults,refertotablesintheSupplementaryInformation, Colon-Perezetal. 2017 ).Aftercorrectingformultiple comparisonsandcontrollingforgraphdensity,statisticallysigni“cantchangeswereobserved for27distinctnodesinnodestrength( Figure3 ),andtwonodesforpathlength( Figure4 ).There werenoobserveddifferencesinclusteringforanynode.Theleft(Lf)parsopercularisandright (Rt)putamenshowedstatisticaldifferencesinnodestrengthandpathlength.Thelocationof thenodeswithstatisticallysigni“cantdifferencescanbeseeninSupplementaryFigure1. BilateraldifferencesinnodestrengthbetweenPD-MIandcontrolswerefoundforthefollowing:putamen( pLf=0.043 and pRt=0.002 ),caudalmiddlefrontal( pLf=0.018 and pRt=0.040 ),inferiorparietal( pLf=0.031 and pRt=0.019 ),postcentral( pLf=0.017 and pRt=0.045 ),posteriorcingulate( pLf=0.043 and pRt=0.006 ),precentral( pLf=0.017 and pRt=0.019 ),andprecuneus( pLf=0.019 and pRt=0.043 ).Additionalnodeswithstatisticallysigni“cantdifferenceswerethefollowing:Lfpallidum( p=0.021 ),Lfentorhinalcortex ( p=0.004 ),Lfisthmuscingulate( p=0.027 ),Lfmiddletemporal( p=0.016 ),Lfparsopercularis( p=0.003 ),Lfbanksofthesuperiortemporalsulcus( p=0.019 ),LfsupramarginalNetworkNeuroscience 114

PAGE 10

ConnectomesinParkinson’sdisease Figure3. Nodestrengthresultsinaconnectomeof82corticalandsubcorticalnodes.Controls ( n=40 ),Parkinsonsparticipantswithoutmemoryimpairment( n=31 ),andParkinsonsparticipantswithmemoryimpairment( n=9 ).Theleftcolumnrepresentsthelefthemispherenodes, andtherightcolumncorrespondstotherighthemisphere.Highlightsingrayrepresentasigni“cant difference(correctedformultiplecomparisons)betweencontrolsandPD-MI.Nosigni“cantdifferenceswereobservedbetweenPD-Wellandcontrols.Bankstemporalsulcusisanabbreviationfor thebanksofthesuperiortemporalsulcusregion,whichisthenameintheFreeSurfernomenclature; thefullnameisusedinthemaintext.( p=0.002 ),Lfrostralmiddlefrontal( p=0.003 ),Lfsuperiortemporal( p=0.006 ),Rtlateraloccipital( p=0.029 ),Rtrostralanteriorcingulate( p=0.019 ),Rtrostralmiddlefrontal ( p=0.007 ),andRtsuperiorparietal( p=0.040 ).ThepathlengthwasdifferentforLfpars opercularis( p=0.002 )andRtputamen( p=0.001 ).Localindicesofnodestrength,clustering coef“cient,andpathlengthdidnotshowanydifferencesbetweencontrolsandPD-Wellor controlsandPD-MIusingFAastheedgeweightaftercorrectingformultiplecomparisons.CorrelationsGiventhelargenumberofcorrelations(threenetworkindices,threecognitive indices,andUPDRSscoresforPDs,82nodes,andthreesubjectgroups),thediscussionis restrictedtothecorrelationanalysisofthosenodesthatshowedstatisticallysigni“cantdifferencesbetweencontrolsandPD-MI.Spearmancorrelationswereperformedfor27nodes(i.e., thosewithstatisticallysigni“cantdifferencesbetweenPD-MIandcontrols; Figure3 ),threeNetworkNeuroscience 115

PAGE 11

ConnectomesinParkinson’sdisease Figure4. Pathlengthresultspernodeinaconnectomeof82corticalandsubcorticalnodes. Controls( n=40 ),Parkinsonsparticipantswithoutmemoryimpairment( n=31 ),andParkinsons participantswithmemoryimpairment( n=9 ).Theleftcolumnrepresentsthelefthemispherenodes, andtherightcolumncorrespondstotherighthemisphere.Highlightsingrayrepresentasigni“cant difference(correctedformultiplecomparisons)betweencontrolsandPD-MI.Nosigni“cantdifferenceswereobservedbetweenPD-Wellandcontrols.Bankstemporalsulcusisanabbreviationfor thebanksofthesuperiortemporalsulcusregion,whichisthenameintheFreeSurfernomenclature; thefullnameisusedinthemaintext.networkindices,andthreeneuropsychologicalindices(PDparticipantsalsoincludeUPDRS scores).Thisanalysisyieldsamaximumof243correlationsincontrolsand324forindividualswithPD(thecorrelationsforallnodescanbefoundintheonlinesupplementaldata; Colon-Perezetal. 2017 ). Controls:Thecontrolgroupdidnotexhibitsigni“cantcorrelationsbetweenanynetwork metric(i.e.,nodestrength,clustering,orpathlength)andanyneuropsychologicalcomposite (workingmemory,memorycomposite,orprocessing-speedcomposite).See Figure5 ,left column. PD-Well:Therewere12correlationswithcoef“cientvalueslargerthan0.5betweennetworkindicesandallneuropsychologicalcomposites(compositesdescribedincognitivemeasures).See Figure5 ,middlecolumn.ForPD-Well,thenodestrengthcorrelatedwiththeNetworkNeuroscience 116

PAGE 12

ConnectomesinParkinson’sdisease Figure5. Thenumberofsigni“cantcorrelations(absolutevalue|r| >0.5 )betweennetwork indicesandeachneuropsychologicalcompositean dUPDRSPartIIIscores(i.e.,motorsymptomsof PD).Theintensityofeachpixelisthenumberofcorrelationslargerthan0.5foreachnetworkmetric andthespeci“ednodewiththeneuropsychologicalcomposites(i.e.,workingmemory,memory composite,speedcomposite,andUPDRSPartIII).Thereisamaximumoffourpossiblesigni“cant correlations,computedbetweenasinglenetworkmetricandthefourneurocognitivemeasures. Thisisthenreportedforeachnodewithsigni“cantdifferencesinnodestrengthbetweenPD-MI andcontrols(see Figure3 ).Bankstemporalsulcusisanabbreviationforthebanksofthesuperior temporalsulcusregion,whichisthenameintheFreeSurfernomenclature;thefullnameisusedin themaintext.compositesforthefollowingnodes:Lfinferiorparietal,Lfparsopercularis,Lfpostcentral,Lf posteriorcingulate,Lfprecentral,Lfsupramarginal,Rtposteriorcingulate,andRtprecentral. Theclusteringcoef“cientdidnotcorrelatewithanycomposite.Thepathlengthcorrelatedfor theLfinferiorparietal,Lfprecentral,Lfpostcentral,andLfsupramarginal. PD-MI:Therewere63correlationswithvalueslargerthan0.5betweennetworkindices andallneuropsychologicalcomposites.Themajorityofnodescorrelatedwithatleastone neuropsychologicalcompositeforthePD-MIgroup( Figure5 );therefore,inthissection,we willlistthenodesthatdidnotshowcorrelationsbetweennetworkindicesandcomposites. ThenodestrengthofPD-MIdidnotcorrelatewithLfbanksofthesuperiortemporalsulcus,Lf caudalmiddlefrontal,Lfentorhinal,Lfrostralmiddlefrontal,Lfsuperiortemporal,Rtputamen, Rtlateraloccipital,andRtsuperiorparietal.Theclusteringcoef“cientdidnotcorrelatewith compositesforLfmiddletemporal,Lfposteriorcingulate,Lfsupramarginal,Rtinferiorparietal, Rtprecentral,Rtsuperiorparietal,andRtposteriorcingulate.Thepathlengthdidnotcorrelate fortheLfentorhinalcortex,Rtputamen,andRtsuperiorparietal.NetworkNeuroscience 117

PAGE 13

ConnectomesinParkinson’sdisease DISCUSSIONInthisstudy,weidenti“eddifferencesintheconnectomeorganizationforpeoplewithidiopathicPDwithmildamnesticdisturbance(PD-MI),andweexploredcorrelationsbetween connectomesandthemostvulnerablecognitivedomainswithinPDrelativetonon-PDpeers. Globaldifferenceswereidenti“edinthePD-MIversuscontrolgroups,inmeannodestrength andpathlengthbutnotinclusteringcoef“cients,small-worldness,orgraphdensity.Nodifferenceswereobservedforanyoftheindices(globalorlocal)forthePD-Wellgroupversus controls.Wefoundthat27outof82nodesshowedlocaldifferencesofconnectivityinnode strengthinthePD-MIgrouprelativetocontrols( Figure3 ).Localnodenetworkdifferencesin nodestrength and pathlengthwerefoundinPD-MIbrainsandwerespeci“ctotheLfpars opercularis,andRtputamen(SupplementaryFigure1).Bilateraldifferencesbetweencontrols andPD-MIwerefoundinnodestrengthfortheputamen,caudalmiddlefrontal,inferiorparietal,postcentral,posteriorcingulate,rostralmiddlefrontal,precentral,andprecuneus.The rightputamenwastheonlysubcorticalregionthatdisplayedasigni“cantconnectomealterationintheformofareductioninnodestrengthandpathlength.Nodestrengthandcognitive compositesfurthershowedapotentiallarge-scalenetworkconnectivityreductioninPD-MI relativetonormalcognitiveareasofvulnerability.Thisstudyshowskeyconnectomeindices forconsiderationinPD-WellandPD-MIphenotypes.StructuralDifferencesRobustconnectomedifferencesbetweenPD-MIandcontrolswereobservedatthegloballevel withmeannodestrengthandmeanpathlength.Inthisstudy,groupsstructuralconnectomes showedmanysimilarlevelsofconnectivity,asshownintheweightedtopedgesin Figure1A andtheirgraphdensitiesin Figure1B .Relativetonon-PDpeers,meannodestrengthwasalteredby5%and13%forPD-WellandPD-MI,respectively,andmeanpathlengthwaschanged 2%and11%,respectively.These“ndingsinnode strengthandpathlen gthsuggestareduction intheintegrityofwhitematterconnectivityinPD.ThepathlengthalterationinPDisdirectly relatedtothenodestrengthreduction(i.e., r=0.92 ).Theedges(i.e.,edgeweights)connectingnodesthatinturnyieldth eshortestpathbetweennodespossiblydecreasesbecauseof theneurodegenerationofwhitematterinPD( Tessitore,Giordano,Russo,&Tedeschi 2016 ). Theabsenceofgroupdifferencesinclusteringcoef“cientandsmall-worldness,whichiscon“rmedbyapreviousreport( Galantuccietal. 2016 ),maysuggestamethodtocompensate forPDchangeswherethebrainnetworkadaptstopreserveitssmall-worldnessandclustering features.Therefore,thechangesinnodestrengthandpathlengthcouldbemarkersofPD progressionandcognitivedecline.Althoughcross-sectionalandpreliminary,these“ndings suggestalocalandglobalnetworkreductioninc onnectivitywithanamnesticdisturbance (i.e.,PD-MI). Tofurtherexploretherelevanceofparticularnodealterations,weperformedlocal topologicalanalysesthatrevealedevidenceoflocalnetworkdisruptioninPD-MI.Node strengthandpathlengthabnormalitieswereseenwiththeleftparsopercularisandright putamen.Thefrontalregionisinvolvedinnetworksoflanguage,attention,andworkingmemory( Lezak 2012 ; Petrides 2000 ; Stussetal. 2005 ; Zola-Morgan&Squire 1986 ; Zola-Morgan,Squire,&Amaral 1986 ),whilethesubcorticalnucleioftheputamenisinvolvedinPDmotorsymptomatology( Braak,Ghebremedhin,Rub,Bratzke,&DelTredici 2004 ; Lisanbyetal. 1993 ; Nemmi,Sabatini,Rascol,&Peran 2015 ; Priceetal. 2016 ).Althoughpreliminary,theseconnectivityreductionsprimarilyinleftcorticalareasvalidatethe pro“leofourPD-MI,particularlygiventhatthePD-MIgroupwasclassi“edwithverbalmemoryNetworkNeuroscience 118

PAGE 14

ConnectomesinParkinson’sdisease measures,whichinvolvelefthemisphereregionsmorethantheright( Golbyetal. 2001 ).This pro“leisalsoconsistentwithotherreportsshowinganteriorandlateraltemporalthinningand volumereductioninPD-MI( Crowleyetal. 2017 ; Pagonabarragaetal. 2013 ; Tanneretal. 2015 ). Atthispoint,wewouldliketoillustratealargerissueintheconnectomicsliterature:thelackofagroundtruthedgeweight.Inthiswork,wefoundadiscrepancybetweenourweightedframework( Colon-Perezetal. 2016 )andanFAweightingscheme. Severalworkshavebeenpublishedthatbringattentiontotheinadequaciesoftraditionallyusededgeweightingschemesinnetworkneuroscience. Chengetal. ( 2012 )“rstdescribedhowincreasingseeddensityimprovesthestabilityofnetworkmetrics.However, thereisacaveatthathigherseeddensitiesleadtoalargernumberofspuriousstreamlines andthusaffecttheconnectome.Inouredgeweightscheme,weemployalargenumberofseedspervoxeltoincreaseitsstabilityandusethecharacteristicfunction(Equation2)tomitigatetheeffectsofspuriousstreamlines( Colon-Perez,Spindler,etal. 2015 ). Buchanan,Pernet,Gorgolewski,Storkey,andBastin ( 2014 )suggestthatsomemeasureof streamlinedensity(asouredgeweight)issuperiortoFAsinceitproducesbettertest-retest performance.Thesebiasesandothersareattemptedtobecontrolledorreducedbyour weightingscheme( Colon-Perez,Spindler,etal. 2015 ).Ourworkandthatofmanyothers areunderwaytodevelopnewandbetterwaystoweighnetworksinconnectomestudiesthat improvethestabilityofnetworkmetricsderivedfromtractography( Colon-Perezetal. 2016 ; Colon-Perez,Spindler,etal. 2015 ; Girard,Whittingstall,Deriche,&Descoteaux 2014 ).We arenotclaimingthatourweightingschemeisa moreaccuraterepresentationoftheunderlying anatomicalconnectivitythanothers;butwithourpreviousarticlesandthisone,wehopeto continuethediscussionto“ndnewandnovelwaystoweighconnectomes. FunctionalMRIstudieshaveshownthatdisruptionsinthemotornetworksofpatientswith PDcorrelatewithdiseaseseverity( Wuetal. 2009 ).Thesefunctionalchangesarelikelyaccompaniedbystructuraldisruption,potentiallyliketheonesdescribedhere.Microstructural changeshavebeenreportedinPDusingdiffusiontensorimaging.LowerFAhasbeenobserved inthesubstantianigraandthestriatuminpatientswithPD( Tessitoreetal. 2016 ).Incontrast, FAincreaseshavebeennotedinthecorpuscallosumandthesuperiorlongitudinalfasciculus( Gattellaroetal. 2009 ).Thiswhitematterchangeleadsustohypothesizethatinterhemisphericconnectivityalterationsaremediatedviathecorpuscallosumandwithin-hemisphere alterationsthroughthesuperiorlongitudinalfasciculus( Northametal. 2012 ).Reductionsin FAhavealsobeenreportedinparticipantswithPDintheputamen,substantianigra,striatum, frontallobes,andmotorareas( Zhanetal. 2012 ).Thesebrainconnectivitychangesarenot onlyrestrictedtoMRIbutalsohavebeenobservedwithSPECT( Booijetal. 1997 )andPET ( Brooks 1995 ).Altogether,theresearch“ndings,in cludingourstudy,suggestmanychanges inPDmaycoalesceintoareorganizationofthestructuralbrainconnectome.Werecognize thattheseconnectomedifferencesmaynotbe uniquetomemory-impairedPDbecausesimi-lardifferenceshavebeenreportedinpatientswithepilepsyafteranteriortemporallobectomy ( Jietal. 2015 ),andinindividualswithPDafterdeepbrainstimulation( vanHarteveltetal. 2014 ).Cognitive-NetworkCorrelationsCognitive-networkanalyseshe lptovalidategroup-levelstruc turaldifferences.Weexamined keycognitivedomainscompromisedinPD:processingspeed,workingmemory,andepisodicNetworkNeuroscience 119

PAGE 15

ConnectomesinParkinson’sdisease memory.OurobservationssuggestthatPD-MIhadmorecorrelationsbetweenneuropsychologicalcompositesandconnectivityindicesthanPD-Wellandnon-PDpeers.ForPD-Well, correlationswerefoundbetweenworkingmemoryandnodestrength,aswellasforworking memoryandpathlength.ThePD-Wellgroup sworkingmemoryshowedanegativecorrelationwithnodestrengthandpathlength(i.e.,highernodestrengthandpathlengthswere associatedwithlowerworking-memoryscore s).Thetopologicalchangessuggestapossible maladaptationinthebrainnetworksofthosewithPD,assuggestedinotherneurologicaldisorders( Doucetetal. 2015 ; Drakesmithetal. 2015 ; Wuetal. 2009 ).Thus,thesetopological indicesmightbeusedasmarkersforcognitiveperformanceinPD. Therewerenosigni“cantlocalnetworkcorrelationsbetweenthenetworkindicesandneuropsychologycompositesinthecontrolgroup( Figure5 ).Thisnegative“ndingisexpected giventhisgroupsrelativecohesiveness,intactperformance,andrestrictedrangeofscores.In contrast,thePD-Wellgroupshowedanincreasednumberofstrongcorrelations,whilethe PD-MIshowedanevenhighernumberofstrongcorrelations.Theincreaseinthenumberof correlationsmightre”ectaprogressivealterationofbrainconnectivitymediatingthecognitive decline.Althoughweobservedincreasingnumbersofcorrelationsbetweennetworkindices andneuropsychologicalcomposites,wedidnot“ndverbalmemorycorrelationstoanyof thenetworkindices.Thelackofcorrelationsbetweenverbalmemoryandnetworkindicesin PD-MIwaspossiblyduetotherestrictedrangeoftheirscores(allimpaired).StudyStrengthsandLimitationsOverall,thestudystrengthsincludethepresentationofacomprehensivecorrelationalanalysis betweennetworktopologyandcognitionusingvariousneuropsychologicalindices,prospectivePDandcontrolmatchingondemographicvariables(i.e.,ageandeducation),androbust identi“cationofmemoryimpairmentinPD.ThePDsubgroupcognitivepro“lesshowedexpectedreductionsinprocessingspeedforPDrelativetonon-PDcontrols,withPD-MIshowingreductionsinmemoryandprocessingspeed.Inthiswork,weusedaweightednetwork methodthatyieldsmorestabletopologicalmetricresultsthanbinarynetworkmethodsandis robustdespitegraphdensitydifferences;hence,itdoesnotrequirethresholdingtogenerate theconnectomes( Colon-Perezetal. 2016 ).Topologicalfeaturesofbinarynetworkconnectomesmaybeaffectedbytheirgraphdensityandultimatelymayhindercomparisonsbetween groups( Langer,Pedroni,&Jancke 2013 ; Sporns 2011b ).Wecouldcircumventtheproblem ofthresholdingbyemployingaweightedframeworkthatreducestheeffectsofthresholdingin networkindicesresults( Colon-Perezetal. 2016 ). Theauthorsrecognizesomeofthestudylimitations.MRtractographyissusceptibleto falsepositivesandfalsenegatives;hence,careisrequiredwhenanalyzingandinterpreting tractography-derivedresults( Alhourani&Richardson 2015 ).Inthisstudy,weusedadifferent edgeweighttocalculatethestrengthofconnectivitybetweennodes.Thisapproachreduces thetractographybiaseffectsofseeddensityandlength( Colon-Perez,Spindler,etal. 2015 ). Also,theedgeweightisde“nedonlyalongthepathofthestreamlines,whichreducesthe possibilityofobtainingstreamlinesfromextraneousareas.Also,theuseofaweightednetworkde“nesconnectomesasasetofweakandstrongconnections,whichhasbeenshown toreducetheeffectsoffalse-positiveconnectionsinthequanti“cationofconnectomeindices usingbinarynetworks( Colon-Perezetal. 2016 ).Moreover,false-positiveconnectionsyield connectomeedgeswithasmallnumberofstreamlines,whichinturncorrespondstoalow edgeweightvalue;thesespuriouslinkswillonlyproducechangesinnodestrengthoflessthan 1%( Colon-Perezetal. 2016 ).Anotherlimitationistherelativelylowspatialresolution(2mmNetworkNeuroscience 120

PAGE 16

ConnectomesinParkinson’sdisease isotropicfordiffusionimages)intractography,whichisalimitingfactorwhenestimatingsmall tractsandmayleadtofalsenegatives( Colon-Perez,King,etal. 2015 ; Fordetal. 2013 ).In thecurrentwork,weusedhighangular-resolutiondiffusionimagingtoincreasetheangular resolutionandenablebetterestimationofthe“berorientationoneachvoxel( Jianetal. 2007 ; Tuchetal. 2002 ).Italsohasbeenreportedthattheconnectomechangeswithage( Wuetal. 2007 ).However,themeanandrangeofageineachgroupissimilartoensuretherewereno agedifferencesbetweengroups.Thus,thecurre ntresultsmayonlyapplytosimilarlyaged populations.Wealsodidnotconsidergenetics,whichhasbeenshowntomodulateconnectometopologyinpatientswithPDthroughthers405509riskallele(TT)( Shuetal. 2015 ).An additionallimitationistherelativelysmallsamplesizeofthisstudy,particularlywithinthe PD-MIgroup( n=9 ).Futurestudiesareencouragedusin glargersamplesizesandadditional explorationsofnetworkdifferencesinmemoryve rsusothercognitive(e .g.,executiveworking, attention)de“cits.CONCLUDINGREMARKSOurconnectomeanalysessuggestalossandreorganizationofbrainwhitematterstructure inPD,particularlyPDwithmemoryimpairment.Weidenti“edareductionofconnectome topologicalandneuropsychologicalindices.Ourresultsshowarelationshipbetweencognitivede“citsandconnectomestructurealterationinPDwithmildcognitiveimpairments. Itremainstobedeterminedwhethertheobservednetworkchangesarecausaloftheneurocognitivede“citsorviceversa.Also,furtherstudiesareneededtoassessthemechanisms relatingtotheobservedtopologicalchangesinbrainstructuretotheneurocognitivede“cits. Thedatasuggestabroaderchange,atthelevel oftheconnectome,associatedwiththeclinical manifestationsofcognitivephenotypes,particularlythememoryphenotypeofPD.ACKNOWLEDGMENTSWearegratefultotheparticipantsinvolvedinthecurrentinvestigation.Wearealsogratefulto TonyMancuso,MD,forhishelpinsecuringtheMRSiemensVerio,andtheUFRadiologyteam fortheirguidance.WewouldliketoacknowledgeIreneMalaty,MD,RamonRodriguez,MD, JanetRomrell,ARNP,PamZeilman,ARNP,andallthefacultyintheUFCenterforMovement DisordersandNeurorestoration,Gainesville,Florida,forreferringindividualstotheinvestigation.WealsothankJessicaWilliams,CassieCatania,JadeWard,KatieRodriguez,andBreana Wallace,fortheirassistancewithparticipantrecruitment.AUTHORCONTRIBUTIONSLuisM.Colon-Perez:Conceptualization;Formalanalysis;Investigation;Methodology;Project administration;Software;Writing…originaldraft;Writing…review&editing.JaredJ.Tanner: Conceptualization;Datacuration;Formalanalysis;Methodology;Writing…originaldraft; Writing…review&editing.MichelleCouret:For malanalysis;Investigation;Writing…review &editing.ShelbyGoicochea:Formalanalysis;Investigation.ThomasH.Mareci:Project administration;Supervision;Conceptualization;Methodology;Writing…review&editing. CatherineC.Price:Conceptualization;Formalanalysis;Fundingacquisition;Investigation; Methodology;Projectadministration;Supervision;Writing…originaldraft;Writing…review &editing.NetworkNeuroscience 121

PAGE 17

ConnectomesinParkinson’sdisease FUNDINGINFORMATIONFundingforthisresearchwassupportedbytheNationalInstitutesofHealth(grantnos.K23 NS060660,RO1NR014181,andR01NS082386),bytheNationalInstitutesofHealthClinicalandTranslationalScienceAwardprogram(grantnos.UL1TR001427,KL2TR001429,and TL1TR001428),andbytheUFCenterforMovementDisordersandNeurorestorationandthe BrainandSpinalCordInjuryResearchTrustFundoftheStateofFlorida.Thecontentissolely theresponsibilityoftheauthorsanddoesnotnece ssarilyrepresentthe of“cialviewsofthe NationalInstitutesofHealth.REFERENCES Alhourani,A.,&Richardson,R.M.(2015).Inherentlimitationsof tractographyforaccuratec onnectivitymaps. Neurosurgery 76 N11…N12. Basser,P.J.,&Jones,D.K.(2002).Diffusion-tensorMRI:Theory, experimentaldesignanddataanalysis-atechnicalreview. NMR inBiomedicine 15 ,456…467. Basser,P.J.,Pajevic,S.,Pierpaoli,C.,Duda,J.,&Aldroubi,A. (2000).Invivo“bertractographyusingDT-MRIdata. Magnetic ResonanceinMedicine 44 ,625…632. Bellucci,A.,Mercuri,N.B.,Venneri,A.,Faustini,G.,Longhena, F.,Pizzi,M.,...Spano,P.(2016).Review:Parkinsonsdisease: Fromsynapticlosstoconnectomedysfunction. Neuropathology andAppliedNeurobiology 42 ,77…94. Boccaletti,S.,Latora,V.,More no,Y.,Chavez,M.,&Hwang,D.U. (2006).Complexnetworks:Structureanddynamics. Physics Reports 424 ,175…308. Booij,J.,Tissingh,G.,Boer,G.J.,Speelman,J.D.,Stoof,J.C., Janssen,A.G.,...vanRoyen,E.A.(1997).[123I]FP-CITSPECT showsapronounceddeclineofstriataldopaminetransporter labellinginearlyandadvancedParkinsonsdisease. Journalof Neurology,Neurosurgery,andPsychiatry 62 ,133…140. Braak,H.,Ghebremedhin,E.,Rub,U.,Bratzke,H.,&DelTredici, K.(2004).St agesinthedevelopmentofParkinsonsdiseaserelatedpathology. CellandTissueResearch 318 ,121…134. Briggs,G.G.,&Nebes,R.D.(1975).Patternsofhandpreference inastudentpopulation. Cortex 11 ,230…238. Brooks,D.J.(1995).Theroleofthebasalgangliainmotorcontrol: ContributionsfromPET. JournaloftheNeurologicalSciences 128 ,1…13. Buchanan,C.R.,Pernet,C.R.,Gorgolewski,K.J.,Storkey,A.J., &Bastin,M.E.(2014).Test-retestreliabilityofstructuralbrain networksfromdiffusionMRI. NeuroImage 86 ,231…243. Catani,M.,&Ffytche,D.H.(2005).Therisesandfallsofdisconnectionsyndromes. Brain 128 ,2224…2239. Charlson,M.E.,Pompei,P.,Ales,K.L.,&MacKenzie,C.R.(1987). Anewmethodofclassifyingprognosticcomorbidityinlongitudinalstudies:Developmentandvalidation. JournalofChronic Diseases 40 ,373…383. Chaudhuri,K.R.,&Schapira,A.H.(2009).Non-motorsymptoms ofParkinsonsdisease:Dopaminergicpathophysiologyandtreatment. TheLancetNeurology 8 ,464…474. Cheng,H.,Wang,Y.,Sheng,J.,Sporns,O.,Kronenberger,W.G., Mathews,V.P.,...Saykin,A.J.( 2012).Optimizationofseed densityinDTItractographyforstructuralnetworks. Journalof NeuroscienceMethods 203,264…272. Cohen,J.(1992).Apowerprimer. PsychologicalBulletin 112 155…159. Colon-Perez,L.M.,Couret,M.,Triplett,W.,Price,C.,&Mareci, T.H.(2016).Smallworldnessindenseandweightedconnectomes. FrontiersinPhysics 4 Colon-Perez,L.M.,King,M.,Parekh,M.,Boutzoukas,A., Carmona,E.,Couret,M.,...Carney,P.R.(2015).High“eldmagneticresonanceimagingofthehumantemporallobe. NeuroImage:Clinical 9 ,58…68. Colon-Perez,L.M.,Spindler,C.,Goicochea,S.,Triplett,W.,Parekh, M.,Montie,E.,...Mareci,T.H.(2015).Dimensionless,scale invariant,edgeweightmetricforthestudyofcomplexstructural networks. PLoSONE 10 ,e0131493. Colon-Perez,L.M.,Tanner,J.J.,Couret,M.,Goicochea,S.,Mareci, T.H.,&Price,C.C.(2017).Supplementalmaterialfor CognitionandconnectomeinnondementiaidiopathicParkinsonsdisease.Ž NetworkNeuroscience 2 (1),106…124. https:// doi.org/10.1162/netn _a_00027 Crowley,S.J.,Huang,H.,Tanner,J.J.,Zho,Q.,Schwab,N.A., Hizel,L.P.,...Price,C.C.(2017).Consideringtotalintracranialvolumeandothernuisancevariablesinbrainvoxel basedmorphometryinidiopathicPD. BrainImagingand Behavior ,1…12. Daianu,M.,Jahanshad,N.,Nir,T.M.,Jack,C.R.,Jr.,Weiner,M.W., Bernstein,M.A.,&Thompson,P.M.(2015).RichclubanalysisintheAlzheimersdiseaseconnectomerevealsarelatively undisturbedstructuralcorenetwork. HumanBrainMapping 36 3087…3103. Doucet,G.E.,Rider,R.,Taylor,N.,Skidmore,C.,Sharan,A., Sperling,M.,&Tracy,J.I.(2015).Presurgeryresting-statelocalgraph-theorymeasurespredictneurocognitiveoutcomesafter brainsurgeryintemporallobeepilepsy. Epilepsia 56 ,517…526. Drakesmith,M.,Caeyenberghs,K.,Dutt,A.,Zammit,S.,Evans, C.J.,Reichenberg,A.,...Jones,D.K.(2015).Schizophrenialiketopologicalchangesinthestructuralconnectomeof individualswithsubclinicalpsychoticexperiences. HumanBrain Mapping 36 ,2629…2643. Erds,P.,&Rnyi,A.(1959).Onrandomgraphs,I. Publicationes Mathematicae(Debrecen) 6 ,290…297. Fling,B.W.,Cohen,R.G.,Mancini,M.,Carpenter,S.D.,Fair,D.A., Nutt,J.G.,&Horak,F.B.(2014).FunctionalreorganizationofNetworkNeuroscience 122

PAGE 18

ConnectomesinParkinson’sdisease thelocomotornetworkinParkinsonpatientswithfreezingofgait. PLoSONE 9 ,e100291. Ford,A.A.,Colon-Perez,L.,Triplett,W.T.,Gullett,J.M.,Mareci, T.H.,&Fitzgerald,D.B.( 2013).Imagingwhitematterinhuman brainstem. FrontiersinHumanNeuroscience 7 ,400. Galantucci,S.,Agosta,F.,Stefanova,E.,Basaia,S.,vandenHeuvel, M.P.,Stojkovic,T.,...Filippi,M.(2016).StructuralbrainconnectomeandcognitiveimpairmentinParkinsondisease. Radiology ,160274. Gattellaro,G.,Minati,L.,Grisoli,M.,Mariani,C.,Carella,F.,Osio, M.,...Bruzzone,M.G.( 2009).Whitematterinvolvementin idiopathicParkinsondisease:Adiffusiontensorimagingstudy. AmericanJournalofNeuroradiology 30 ,1222…1226. Genovese,C.,&Wasserman,L.(2002).Operatingcharacteristics andextensionsofthefalsediscoveryrateprocedure. Journalof theRoyalStatisticalSociety:Ser iesB(StatisticalMethodology) 64 ,499…517. Girard,G.,Whittingstall,K.,Deriche,R.,&Descoteaux,M.(2014). Towardsquantitativeconnectivityanalysis:Reducingtractographybiases. NeuroImage 98 ,266…278. Golby,A.J.,Poldrack,R.A.,Brewer,J.B.,Spencer,D.,Desmond, J.E.,Aron,A.P.,&Gabrieli,J.D.(2001).Material-speci“clateralizationinthemedialtemporallobeandprefrontalcortexduring memoryencoding. Brain 124 ,1841…1854. Golden,C.J.,&Freshwater,S.M.(2002). TheStroopColorand WordTest:Amanualforclinicalandexperimentaluses .Chicago, IL:Stoelting. Granovetter,M.S.(1973).Thestrengthofweakties. American JournalofSociology 78 ,1360…1380. Gratwicke,J.,Jahanshahi,M.,&Foltynie,T.(2015).Parkinsons diseasedementia:Aneuralnetworksperspective. Brain 138 1454…1476. Grenander,U.(1956).Onthetheoryofmortalitymeasurement. ScandinavianActuarialJournal 1956 ,125…153. Hagmann,P.,Cammoun,L.,Gigandet,X.,Gerhard,S.,EllenGrant, P.,Wedeen,V.,...Sporns,O.(2010).MRconnectomics:Principlesandchallenges. JournalofNeuroscienceMethods 194 34…45. Heaton,R.K.,&PsychologicalAssessmentResourcesInc.(2004). RevisedcomprehensivenormsforanexpandedHalstead-Reitan battery:Demographicallyadjustedneuropsychologicalnorms forAfricanAmericanandCaucasianadults,professionalmanual [Updateded.].Lutz,FL:PsychologicalAssessmentResources. Henderson,E.J.,Chu,H.,Gaunt,D.M.,Whone,A.L.,BenShlomo,Y.,&Lyell,V.(2016).ComparisonofTestYourMemory andMontrealCognitiveAssessmentmeasuresinParkinsonsdisease. ParkinsonsDisease 2016 ,1012847. Hoehn,M.M.,&Yahr,M.D.(1967).Parkinsonism:Onset,progressionandmortality. Neurology 17 ,427…442. Hughes,A.J.,Ben-Shlomo,Y.,Daniel,S.E.,&Lees,A.J.( 2001). Whatfeaturesimprovetheaccuracyofclinicaldiagnosisin Parkinsonsdisease:Aclinicopathologicstudy.1992.Neurology 57 ,S34…38. Humphries,M.D.,&Gurney,K.(2008).Networksmall-worldnessŽ:Aquantitativemethodfordeterminingcanonicalnetwork equivalence. PLoSONE 3 ,e0002051. Janvin,C.C.,Larsen,J.P.,Aarsland,D.,&Hugdahl,K.(2006).SubtypesofmildcognitiveimpairmentinParkinsonsdisease:Progressiontodementia. MovementDisorders 21 ,1343…1349. Jenkinson,M.,Beckmann,C.F.,Behrens,T.E.,Woolrich,M.W.,& Smith,S.M.(2012).Fsl. NeuroImage 62 ,782…790. Ji,G.J.,Zhang,Z.,Xu,Q.,Wei,W.,Wang,J.,Wang,Z.,...Lu, G.(2015).Connectomereorganizationassociatedwithsurgical outcomeintemporallobeepilepsy. Medicine(Baltimore) 94 e1737. Jian,B.,Vemuri,B.C.,Ozarslan,E.,Carney,P.R.,&Mareci, T.H.(2007).AnoveltensordistributionmodelforthediffusionweightedMRsignal. NeuroImage 37 ,164…176. Langer,N.,Pedroni,A.,&Jancke,L.(2013).Theproblemofthresholdinginsmall-worldnetworkanalysis. PLoSONE 8 ,e53199. Lezak,M.D.( 2012). Neuropsychologicalassessment (5thed.). NewYork,NY:OxfordUniversityPress. Lisanby,S.H.,McDonald,W.M.,Massey,E.W.,Doraiswamy, P.M.,Rozear,M.,Boyko,O.B. ,...Nemeroff,C.( 1993).DiminishedsubcorticalnucleivolumesinParkinsonsdiseaseby MRimaging. JournalofNeuralTransmission 40 (Suppl.),13…21. Luo,C.Y.,Guo,X.Y.,Song,W.,Chen,Q.,Cao,B.,Yang,J.,... Shang,H.F.(2015).Functionalconnectomeassessedusing graphtheoryindrug-naiveParkinsonsdisease. JournalofNeurology 262 ,1557…1567. Matteau,E.,Dupre,N.,Langlois,M.,Jean,L.,Thivierge,S., Provencher,P.,&Simard,M.(2011).MattisDementiaRating Scale2:ScreeningforMCIanddementia. AmericanJournalof AlzheimersDiseaseandOtherDementias 26 ,389…398. Meijering,E.H.W.,Zuiderveld,K.J.,&Viergever,M.A.(1999).Imagereconstructionbyconvolutionwithsymmetricalpiecewise nth-orderpolynomialkernels. IEEETransactionsonImageProcessing 8 ,192…201. Nemmi,F.,Sabatini,U.,Rascol,O.,&Peran,P.(2015).Parkinsonsdiseaseandlocalatrophyinsubcorticalnuclei:Insightfrom shapeanalysis. NeurobiologyofAging 36 ,424…433. Newman,M.E.J.(2001).Scienti“ccollaborationnetworks.II. Shortestpaths,weightednetworks,andcentrality. PhysicalReviewE 64 Northam,G.B.,Liegeois,F.,Tournier,J.D.,Croft,L.J.,J ohns, P.N.,Chong,W.K. ,...Baldeweg,T.( 2012).Interhemispherictemporallobeconnectivitypredictslanguageimpairmentinadolescentsbornpreterm. Brain 135 ,3781…3798. Odish,O.F.,Caeyenberghs,K.,Hosseini,H.,vandenBogaard, S.J.,Roos,R.A.,&Leemans,A.( 2015).DynamicsoftheconnectomeinHuntingtonsdisease:AlongitudinaldiffusionMRI study. NeuroImage:Clinical 9 ,32…43. Pagonabarraga,J.,Corcuera-Solano,I.,Vives-Gilabert,Y.,Llebaria, G.,Garcia-Sanchez,C.,Pascual-Sedano,B.,...Gomez-Anson, B.(2013).Patternofregionalcorticalthinningassociatedwith cognitivedeteriorationinParkinsonsdisease. PLoSONE 8 e54980. Parkinson,J.(2002).Anessayontheshakingpalsy.1817. Journalof NeuropsychiatryandClinicalNeurosciences 14 ,223…236;discussion222. Petrides,M.(2000).Dissociablerolesofmid-dorsolateralprefrontal andanteriorinferotemporalcortexinvisualworkingmemory. JournalofNeuroscience 20 ,7496…7503.NetworkNeuroscience 123

PAGE 19

ConnectomesinParkinson’sdisease Price,C.C.,Garrett,K.D.,Jefferson,A.L.,Cosentino,S.,Tanner, J.J.,Penney,D.L.,...Li bon,D.J.(2009).Leukoaraiosisseverityandlist-learningindementia. ClinicalNeuropsychologist 23 944…961. Price,C.C.,Tanner,J.,Nguyen,P.T.,Schwab,N.A.,Mitchell, S.,Slonena,E.,...Bowers,D.(2016).Grayandwhitematter contributionstocognitivefrontostriatalde“citsinnon-demented Parkinsonsdisease. PLoSONE 11 ,e0147332. Schwab,N.A.,Tanner,J.J.,Nguyen,P.T.,Schmalfuss,I.M., Bowers,D.,Okun,M.,&Price,C.C.(2015).Proofofprinciple:Transformationapproachalterscaudatenucleusvolumeand structure-functionassociations. BrainImagingandBehavior 9 744…753. Shu,N.,Li,X.,Ma,C.,Zhang,J.,Chen,K.,Liang,Y.,...Zhang, Z.(2015).EffectsofAPOEpromoterpolymorphismonthetopologicalorganizationofbrainstructuralconnectomeinnondementedelderly. HumanBrainMapping 36 ,4847…4858. Sporns,O.(2011a).Thehumanconnectome:Acomplexnetwork. AnnalsoftheNewYorkAcademyofScienes 1224 ,109…125. Sporns,O.(2011b). Networksofthebrain .Cambridge,MA:MIT Press. Strimmer,K.(2008).Fdrtool:AversatileRpac kageforestimating localandtailarea-basedfalsediscoveryrates. Bioinformatics 24 ,1461…1462. Stuss,D.T.,Alexander,M.P.,Shallice,T.,Picton,T.W., Binns,M.A.,Macdonald,R.,...Katz,D.I.(2005).Multiple frontalsystemscontrollingresponsespeed. Neuropsychologia 43 ,396…417. Tanner,J.J.,Levy,S.A.,Schwab,N.A.,Hizel,L.P.,Nguyen,P.T., Okun,M.S.,&Price,C.C.(2017).Markedbrainasymmetrywith intactcognitivefunctioninginidiopathicParkinsonsdisease:A longitudinalanalysis. ClinicalNeuropsychologist 31 ,654…675. Tanner,J.J.,Mareci,T.H.,Okun,M.S.,Bowers,D.,Libon,D.J., &Price,C.C.(2015).Temporallobeandfrontal-subcortical dissociationsinnon-dementedParkinsonsdiseasewithverbal memoryimpairment. PLoSONE 10 ,e0133792. Tanner,J.J.,McFarland,N.R.,&Price,C.C.(2017).Striataland hippocampalatrophyinidiopathicParkinsonsdiseasepatients withoutdementia:Amorphometricanalysis. FrontiersinNeurology 8 ,139. Taylor,P.N.,Han,C.E.,Schoene-Bake,J.C.,Weber,B.,&Kaiser, M.(2015).Structuralconnectivitychangesintemporallobe epilepsy:Spatialfeaturescontributemorethantopologicalmeasures. NeuroImage:Clinical 8 ,322…328. Team,R.C.(2015). R:Alanguageandenvironmentforstatisticalcomputing .Vienna,Austria:RFoundationforStatistical Computing. Tessitore,A.,Giordano,A.,Russo,A.,&Tedeschi,G.(2016).StructuralconnectivityinParkinsonsdisease. ParkinsonismandRelatedDisorders 22 (Suppl.1),S56…S59. Tomlinson,C.L.,Stowe,R.,Patel,S.,Rick,C.,Gray,R.,& Clarke,C.E.(2010).SystematicreviewoflevodopadoseequivalencyreportinginParkinsonsdisease. MovementDisorders ,25 2649…2653. Tuch,D.S.,Reese,T.G.,Wiegell,M.R.,Makris,N.,Belliveau, J.W.,&Wedeen,V.J.(2002).Highangularresolutiondiffusionimagingrevealsintravoxelwhitematter“berheterogeneity. MagneticResonanceinMedicine 48 ,577…582. vanHartevelt,T.J.,Cabral,J.,Deco,G.,Moller,A.,Green,A.L., Aziz,T.Z.,&Kringelbach,M.L.(2014).Neuralplasticityinhumanbrainconnectivity:TheeffectsoflongtermdeepbrainstimulationofthesubthalamicnucleusinParkinsonsdisease. PLoS ONE 9 ,e86496. vanWijk,B.C.,Stam,C.J.,&Daffertshofer,A.(2010).Comparing brainnetworksofdifferentsizeandconnectivitydensityusing graphtheory. PLoSONE 5 ,e13701. Watts,D.J.,&Strogatz,S.H.(1998).Collectivedynamicsof small-worldŽnetworks. Nature 393 ,440…442. Wechsler,D.(1991). WechslerMemoryScale (3rded.).San Antonio,TX:PsychologicalCorporation. Wu,T.,Wang,L.,Chen,Y.,Zhao,C.,Li,K.,&Chan,P.(2009). Changesoffunctionalconnectivityofthemotornetworkinthe restingstateinParkinsonsdisease. NeuroscienceLetters 460 6…10. Wu,T.,Zang,Y.,Wang,L.,Long,X.,Hallett,M.,Chen,Y.,... Chan,P.(2007).Agingin”uenceonfunctionalconnectivityof themotornetworkintherestingstate. NeuroscienceLetters 422 164…168. Xia,M.,Wang,J.,&He,Y.(2013).BrainNetViewer:Anetwork visualizationtoolforhumanbrainconnectomics. PLoSONE 8 e68910. Yendiki,A.,Koldewyn,K.,Kakunoori,S.,Kanwisher,N.,&Fischl, B.(2013).Spuriousgroupdifferencesduetoheadmotionina diffusionMRIstudy. NeuroImage 88c ,79…90. Zgaljardic,D.J.,Borod,J.C.,Foldi,N.S.,&Mattis,P.(2003).A reviewofthecognitiveandbehavioralsequelaeofParkinsons disease:Relationshiptofrontostriatalcircuitry. Cognitiveand BehavioralNeurology 16 ,193…210. Zhan,W.,Kang,G.A.,Glass,G.A.,Zhang,Y.,Shirley,C.,Millin, R.,...Schuff,N.(2012).RegionalalterationsofbrainmicrostructureinParkinsonsdiseaseusingdiffusiontensorimaging. MovementDisorders 27 ,90…97. Zhang,B.,&Horvath,S.(2005).Ageneralframeworkforweighted geneco-expressionnetworkanalysis. StatisticalApplicationsin GeneticsandMolecularBiology 4 ,Article17. Zola-Morgan,S.,&Squire,L.R.(1986).Memoryimpairmentin monkeysfollowinglesionslimitedtothehi ppocampus. BehavioralNeuroscience 100 ,155…160. Zola-Morgan,S.,Squire,L.R.,&Amaral,D.G.(1986).Human amnesiaandthemedialtemporalregion:Enduringmemory impairmentfollowingabilaterallesionlimitedto“eldCA1of thehippocampus. JournalofNeuroscience 6,2950…2967.NetworkNeuroscience 124