I was trying to use ELMER package with some different TCGA tumor types in order to study enhancers. I had no problems with the test example (LUSC), but when I use another TCGA cancer type (GBM) with the default pipeline TCGA.pipe I get the following error:
Error in (function (mee, probes, nearGenes, percentage = 0.2, permu.size = 10000, : Probes option should be subset of row names of methylation matrix.
This error occurs at the beginning of the "Predict pairs" stage. I was checking the results and I think that the error comes from the differential methylation step, so the result files only contains NA values. Someone had this problem?
I was checking also the samples used and I saw that for GBM, the pipeline found 64 tumor samples and 1 control with methylation and expression information, maybe this could be the problem?
Thank you very much.