Package`HIMA'March7,2018TypePackageTitleHigh-DimensionalMediationAnaly - PDF document

Package`HIMA'March7,2018TypePackageTitle
Package`HIMA'March7,2018TypePackageTitleHigh-DimensionalMediationAnalysisVersion1.0.7Date2018-03-06DescriptionAllowstoestimateandtesthigh-dimensionalmediationeffectsbasedonsureindepen-dentscreeningandminimaxconcavepenaltytechniques.Ajointsignicancetestisusedforme-diationeffect.HaixiangZhang,Yi-nanZheng,ZhouZhang,TaoGao,BrianJoyce,GraceYoon,WeiZhang,JoelSchwartz,Al-lanJust,ElenaColicino,PantelVokonas,LihuiZhao,JinchiLv,AndreaBaccarelli,Li-fangHou,LeiLiu(2016) oi:;.1;“/;ioi;&#xnfor;&#xmati;s/b;&#xtw35;ကLicenseGPL-3DependsR oi:;.1;“/;ioi;&#xnfor;&#xmati;s/b;&#xtw35;က(=3.3),ncvregImportsstats,iterators,parallel,foreach,doParallelCollateutils.RsimHIMA.Rhima.RHIMA-package.RLazyDatafalseURLhttps://github.com/YinanZheng/HIMABugReportshttps://github.com/YinanZheng/HIMA/issuesRoxygenNote6.0.1NeedsCompilationnoAuthorYinanZheng[aut,cre],HaixiangZhang[aut],LifangHou[aut],LeiLiu[aut,cph]MaintainerYinanZheng&#xy-zh;ng@;&#xnort;&#xhwes;&#xtern;&#x.edu;RepositoryCRANDate/Publication2018-03-0623:19:18UTC12himaRtop

icsdocumented:HIMA-package..............
icsdocumented:HIMA-package.......................................2hima.............................................2simHIMA..........................................4Index6HIMA-packageHigh-dimensionalMediationAnalysisDescriptionHIMAisanRpackageforestimatingandtestinghigh-dimensionalmediationeffectsingenomic/epigenomicstudies.Package:HIMAType:PackageVersion:1.0.7Date:2018-03-06License:GPL-3Author(s)YinanZheng&#xy-zh;ng@;&#xnort;&#xhwes;&#xtern;&#x.edu;,HaixiangZhang&#xhaix;&#xiang;&#x.zha;&#xng@t;&#xju.e; u.c;&#xn000;,LifangHou&#xl-ho;&#xu@no;&#xrthw;ste;&#xrn.e; u00;Leiliu&#xlei.;&#xliu@;&#xwust;&#xl.ed;&#xu000;Maintainer:YinanZheng&#xy-zh;ng@;&#xnort;&#xhwes;&#xtern;&#x.edu;ReferencesZhangH,ZhengY,ZhangZ,GaoT,JoyceB,YoonG,ZhangW,SchwartzJ,JustA,ColicinoE,VokonasP,ZhaoL,LvJ,BaccarelliA,HouL,LiuL.EstimatingandTestingHigh-dimensionalMediationEffectsinEpigeneticStudies.Bioinformatics.2016.DOI:http://dx.doi.org/10.1093/bioinformatics/btw351.PubMedPMID:27357171.himaHigh-dimensionalMediationAnalysisDescriptionhimaisusedtoestimateandte

sthigh-dimensionalmediationeffects.Usage
sthigh-dimensionalmediationeffects.Usagehima(X,Y,M,COV.XM=NULL,COV.MY=COV.XM,family=c("gaussian","binomial"),penalty=c("MCP","SCAD","lasso"),topN=NULL,parallel=FALSE,ncore=1,verbose=FALSE,...)hima3ArgumentsXavectorofexposure.Yavectorofoutcome.Canbeeithercontinuousorbinary(0-1).Madata.frameormatrixofhigh-dimensionalmediators.Rowsrepresentsam-ples,columnsrepresentvariables.COV.XMadata.frameormatrixofcovariatesdatasetfortestingtheassociationM~X.Covariatesspeciedherewillnotparticipatepenalization.Default=NULL.Ifthecovariatescontainmixeddatatypes,pleasemakesureallcategoricalvariablesareproperlyformattedasfactortype.COV.MYadata.frameormatrixofcovariatesdatasetfortestingtheassociationY~M.Covariatesspeciedherewillnotparticipatepenalization.Ifnotspecied,thesamesetofcovariatesforM~Xwillbeapplied.Usingdifferentsetsofcovari-atesisallowedbutthisneedstobehandledcarefully.familyeither'gaussian'or'binomial',dependingonthedatatypeofoutcome(Y).Seencvregpenaltythepenaltytobeappliedtothemodel.Either'MCP'(thedefault),'SCAD',or'lasso'.Seencvreg.topNanintegers

pecifyingthenumberoftopmarkersfromsurein
pecifyingthenumberoftopmarkersfromsureindependentscreen-ing.Default=NULL.IfNULL,topNwillbeeitherceiling(n/log(n))iffamily='gaussian',orceiling(n/(2*log(n)))iffamily='binomial',wherenisthesamplesize.IfthesamplesizeisgreaterthantopN(pre-speciedorcalculated),allmediatorswillbeincludedinthetest(i.e.low-dimensionalscenario).parallellogical.Enableparallelcomputingfeature?Default=TRUE.ncorenumberofcorestorunparallelcomputingValidwhenparallel==TRUE.Bydefaultmaxnumberofcoresavailableinthemachinewillbeutilized.verboselogical.Shouldthefunctionbeverbose?Default=FALSE....otherargumentspassedtoncvreg.ValueAdata.framecontainingmediationtestingresultsofselectedmediators.•alpha:coefcientestimatesofexposure(X)�–mediators(M).•beta:coefcientestimatesofmediators(M)�–outcome(Y)(adjustedforexposure).•gamma:coefcientestimatesofexposure(X)�–outcome(Y)(totaleffect).•alpha*beta:mediationeffect.•%totaleffect:alpha*beta/gamma.Percentageofthemediationeffectoutofthetotaleffect.•adjusted.p:statistical

signicanceofthemediator(Bonferronipr
signicanceofthemediator(Bonferroniprocedure).•BH.FDR:statisticalsignicanceofthemediator(Benjamini-Hochbergprocedure).4simHIMAExamplesn100#samplesizep500#thedimensionofcovariates#theregressioncoefficientsalpha(exposure&#x--50;�--mediators)alpharep(0,p)#theregressioncoefficientsbeta(mediators&#x--50;�--outcome)beta1rep(0,p)#forcontinuousoutcomebeta2rep(0,p)#forbinaryoutcome#thefirstfourmarkersaretruemediatorsalpha[1:4]c(0.45,0.5,0.6,0.7)beta1[1:4]c(0.55,0.6,0.65,0.7)beta2[1:4]c(1.45,1.5,1.55,1.6)#thesearenottruemediatorsalpha[7:8]0.5beta1[5:6]0.8beta2[5:6]1.7#Generatesimulationdatasimdat_cont=simHIMA(n,p,alpha,beta1,seed=1029)simdat_bin=simHIMA(n,p,alpha,beta2,binaryOutcome=TRUE,seed=1029)#RunHIMAwithMCPpenaltybydefault#WhenYiscontinuous(default)hima.fithima(simdat_cont$X,simdat_cont$Y,simdat_cont$M,verbose=TRUE)head(hima.fit)#WhenYisbinary(shouldspecifyfamily)hima.logistic.fithima(simdat_bin$X,simdat_bin$Y,simdat_bin$M,family="binomial",verbose=TRUE)head(hima.logistic.fit)simHIMASimulationDataGeneratorforHigh-dimensionalMe

diationAnalyaisDescriptionsimHIMAisused
diationAnalyaisDescriptionsimHIMAisusedtogeneratesimulationdataforhigh-dimensionalmediationanalysis.UsagesimHIMA(n,p,alpha,beta,binaryOutcome=FALSE,seed)simHIMA5Argumentsnanintegerspecifyingsamplesize.panintegerspecifyingthedimensionofmediators.alphaanumericvectorspecifyingtheregressioncoefcientsalpha(exposure�–me-diators).betaanumericvectorspecifyingtheregressioncoefcientsbeta(mediators�–out-come).binaryOutcomelogical.Shouldthesimulatedoutcomevariablebebinary?seedanintegerspecifyingaseedforrandomnumbergeneration.SeeAlsoseehimatorunHIMA.Examplesn100#samplesizep500#thedimensionofcovariates#theregressioncoefficientsalpha(exposure&#x--50;�--mediators)alpharep(0,p)#theregressioncoefficientsbeta(mediators&#x--50;�--outcome)betarep(0,p)#thefirstfourmarkersaretruemediators.alpha[1:4]c(0.45,0.5,0.55,0.6)beta[1:4]c(0.5,0.45,0.4,0.35)alpha[7:8]0.5beta[5:6]0.5#Generatesimulationdatasimdat=simHIMA(n,p,alpha,beta,seed=1029)IndexTopicpackageHIMA-package,2HIMA(HIMA-package),2hima,2,5HIMA-package,2ncvreg,3simHIMA,