I was having problems with intraspotCorrelation for a single channel analysis, so I decided to reorganize the data to use duplicateCorrelation. I thought I would get the same answer, but I did not. I hope someone can tell me why. (As you will see below, I like the answer with duplicateCorrelation better!) MAp is the normalized 2-channel data. I have a targets list called targetp with the treatment information. The code is below. ############################################################## # separate channel approach ############################################################# targetSingle <- targetsA2C(targetp) rep=c(0,0,1,1,0,0,1,1,0,0,1,1) designp=model.matrix(~-1+targetSingle$Target+rep,levels=unique(c(cy3,c y5))) corfit =intraspotCorrelation(MAp, designp) corfit$cor NULL ########################################################## # the warning was #In remlscore(y, X, Z) : reml: Max iterations exceeded ########################################################## ############################################################ # approach treating channels as separate arrays ############################################################ RGp=RG.MA(MAp) Rp=log2(RGp$R) Gp=log2(RGp$G) exprP=cbind(Rp[,1],Gp[,1],Rp[,2],Gp[,2],Rp[,3],Gp[,3],Rp[,4],Gp[,4],Rp [,5],Gp[,5],Rp[,6],Gp[,6]) corfitP = duplicateCorrelation(exprP, designp, block = factor(c(1,1,2,2,3,3,4,4,5,5,6,6))) corfitP$cor [1] 0.4921254 sessionInfo() R version 2.11.1 (2010-05-31) i386-pc-mingw32 locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 LC_MONETARY=English_United States.1252 [4] LC_NUMERIC=C LC_TIME=English_United States.1252 attached base packages: [1] tools stats graphics grDevices utils datasets methods base other attached packages: [1] statmod_1.4.8 rat2302cdf_2.6.0 limma_3.4.5 affy_1.26.1 Biobase_2.8.0 loaded via a namespace (and not attached): [1] affyio_1.16.0 preprocessCore_1.10.0 Thanks, Naomi