\name{ClusterPhenoTest} \alias{ClusterPhenoTest} \title{ Test association of clusters with phenotype. } \description{ Test the associations between clusters that each sample belongs to (based on gene expression) and each phenotype. } \usage{ ClusterPhenoTest(x,cluster,vars2test,B=10^4,p.adjust.method='none') } \arguments{ \item{x}{ExpressionSet with phenotype information stored in \code{pData(x)}.} \item{cluster}{variable of class \code{character} or \code{factor} telling at which cluster each sample belongs to.} \item{vars2test}{ list with components 'continuous', 'categorical', 'ordinal' and 'survival' indicating which phenotype variables should be tested. 'continuous', 'categorical' and 'ordinal' must be character vectors, 'survival' a matrix with columns named 'time' and 'event'. The names must match names in \code{names(pData(x))}.} \item{B}{An integer specifying the number of replicates used in the chi-square Monte Carlo test (passed on to \code{chisq.test}).} \item{p.adjust.method}{Method for P-value adjustment, passed on to \code{p.adjust}.} } \details{ Test association between the provided clusters and each phenotype. For variables in vars2test\$continuous and vars2test\$ordinal a Kruskal-Wallis Rank Sum test is used; for vars2test\$categorical a chi-square test (with exact p-value if \code{simulate.p.value} is set to TRUE); for var2test\$survival a Cox proportional hazards likelihood-ratio test. } \examples{ #load data data(eset) eset #construct vars2test survival <- matrix(c("Relapse","Months2Relapse"),ncol=2,byrow=TRUE) colnames(survival) <- c('event','time') #add positive to have more than one category pData(eset)[1:20,'lymph.node.status'] <- 'positive' vars2test <- list(survival=survival,categorical='lymph.node.status') vars2test #first half of the samples will be one cluster and the rest the other cluster cluster <- c(rep('Cluster1',floor(ncol(eset)/2)),rep('Cluster2',ncol(eset)-floor(ncol(eset)/2))) #test association ClusterPhenoTest(eset,cluster,vars2test=vars2test) } \author{ David Rossell }