Dear Bioconductors, I have a problem using maanova. I try to reproduce the analysis in the vignette chapter 5.3 ("An affymetrix experiment") using my own data. I have a time series experiment of developmental stages and I want to detect genes that change their expression over time. The data are generated using Affymetrix chips and the expression set is calculated by RMA. I generate a file of expression values readable by maanova like this: >write.exprs(eset, file="expression.txt") and according to the vignette I read in the data like this: >data.raw<-read.madata("expression.txt", ID=1, pmt=2, spotflag=FALSE) at this point I receive a warning that I ignore, since for affymetrix chips I do not have row and column information: Warning messages: 1: No meta row information, use 1 instead! in: read.madata("expression.txt", ID = 1, pmt = 2, spotflag = FALSE) 2: No meta column information, use 1 instead! in: read.madata("expression.txt", ID = 1, pmt = 2, spotflag = FALSE) 3: No row information. in: read.madata("expression.txt", ID = 1, pmt = 2, spotflag = FALSE) 4: No column information. in: read.madata("expression.txt", ID = 1, pmt = 2, spotflag = FALSE) >data.maanova<-createData(data.raw, 1, log.trans=F) >model.full.mix<-makeModel(data=data.maanova, formula=~Stage+Sample, random=~Sample) >anova.full.mix<-fitmaanova(data.maanova, model.full.mix) in the next step, which is the F-test, I get an error: > test.Stage.mix<-matest(data.maanova, model.full.mix, term="Stage", n.perm=100) Doing F-test on observed data ... Error in var(log(Chis)) : missing observations in cov/cor In addition: Warning message: NAs produced in: rchisq(n, df) My question is: what is wrong here? Where do the missing observations and the NAs come from? As far as I know, my original expression set does not have NAs. How can I find missing observations or NAs in a dataset? And more general: How can I circumvent this problem? I am running the devel version of maanova: Package: maanova Version: 0.98.8 Built: R 2.1.1; powerpc-apple-darwin7.9.0; 2005-09-14 15:01:52; unix Thanks a lot for your help! Georg