预览加载中，请您耐心等待几秒...

基于局部强化非负矩阵分解的基因表达数据聚类.docx

预览
1/3
2/3
3/3

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于局部强化非负矩阵分解的基因表达数据聚类 Abstract Non-negativematrixfactorization(NMF)hasbeenwidelyusedingeneexpressiondataclustering.However,traditionalNMFmethodslacktheabilitytocapturelocalstructuresofthedata,whichcanleadtolessaccurateclusteringresults.Inthispaper,weproposeanovelmethodcalledlocalenhancednon-negativematrixfactorization(LENMF)toaddressthisissue.Specifically,theproposedLENMFmethodenhancesthelocalstructuresofgeneexpressiondatabyaddingaregularizationtermtothecostfunctionofNMF.ExperimentalresultsshowthattheproposedLENMFmethodcanachievebetterclusteringresultsthantraditionalNMFmethods. Introduction Geneexpressiondataanalysisplaysanimportantroleinunderstandingthemechanismofgeneexpressionandidentifyingbiomarkersfordiseases.Clusteringisawidelyusedmethodtogroupgenesorsampleswithsimilarexpressionpatterns.Non-negativematrixfactorization(NMF)hasbeenwidelyusedingeneexpressiondataclusteringduetoitsabilitytodiscovertheunderlyingpatternsinthedata.However,traditionalNMFmethodsignorethelocalstructuresofthedata,whichcanleadtolessaccurateclusteringresults. Toaddressthisissue,manymethodshavebeenproposedtoenhancetheclusteringperformanceofNMF.Forexample,sparseNMF[1]incorporatessparsityconstraintsintoNMFtoselecttherelevantgenesforclustering.ConvexNMF[2]introducesaconvexconstraintintoNMFtocapturetheconvexstructureofthedata.StructuredNMF[3]utilizespriorknowledgeofthegeneinteractionnetworktoimprovetheclusteringresults. Inthispaper,weproposeanovelmethodcalledlocalenhancednon-negativematrixfactorization(LENMF)toenhancethelocalstructuresofthegeneexpressiondata.TheproposedmethodaddsaregularizationtermintothecostfunctionofNMFtoencouragethelearnedbasisandcoefficientmatricestocapturethelocalstructuresofthedata.ExperimentalresultsonthreebenchmarkdatasetsshowthattheproposedLENMFmethodcanachievebetterclusteringresultsthantraditionalNMFmethods. RelatedWork Non-negativematrixfactorization(NMF)hasbeenwidelyusedingeneexpressiondataclustering.GivenageneexpressiondatamatrixX∈Rm×n,NMFfactorizesXintotwonon-negati