Differential Expression Analysis With Affy/Limma On Subset Of Samples?

Question

[bbio]

I want to do differential gene expression analysis on a dataset contained in an AffyBatch object. There are about 500 samples in this set in total, but I only know my variable of interest for a few of them (~15).

a) Is it even realistic to do this on only ~7 vs ~7 samples? b) How can I subset the large dataset to only include the samples which I actually know the group for? I tried setting the design matrix to NA for these, but that doesn't seem to work.

Thanks

[Devon Ryan]

Isn't this what the get.array.subset.affybatch function in simpleaffy does?