Dear Bioconductees, I have been using R for quite some time but am new to both array data and bioconductor (a quantitative geneticist turned molecular geneticist?) I have a question that concerns analyzing arrays that contain replicate spots using limma. I am having trouble with the function dupcor that calculates the correlations between the replicate spots. My replicate spots are not near one another on the physical array ? when I printed the array I printed my 25 print plates containing my probes twice, I wanted to two replicate spots not to be next to each other on the array to reduce spatial correlations. Therefore my replicates are within the same block on the array but not right near each other. There are two ways I can associate the replicate spots. First, there is a ?Name? column in my gal file that includes the specific probe names (i.e. each name appears exactly twice in the gal file). Second, the replicate spots always appear 200 lines apart in the gal file. I thought that a call to duplicateCorrelation with (spacing = 200) I might get things to work, but I get an error message. Here is a little bit of code and the error: >RG <- read.maimages(targets, source = "genepix", wt.fun = rm.flags) >design <- modelMatrix(targets, ref = "Ctl") >total.data <- normalizeWithinArrays(RG) >corfit <- duplicateCorrelation(total.data, design, ndups=2, spacing=200) Error in dim(M) <- c(spacing, ndups, ngroups, nslides) : dims [product 480000] do not match the length of object [483840] So it seems like I am doing something wrong ? An example spot: Gene X is in block 1, row 1, col 1 block 1, row 10, col 12 Gene Y is in block 1, row 1, col 2 block 1, row 10, col 13 Any ideas would be greatly appreciated. Thank you for your time, Kevin -- ========== ========== Kevin J Emerson Bradshaw-Holzapfel Lab Center for Ecology and Evolutionary Biology 1210 University of Oregon Eugene, Oregon 97403 http://www.uoregon.edu/~kemerson kemerson at uoregon.edu