Hello, I am having a strange error message when calling backgroundCorrect with normexp in limma: Error in if (all(abs(delta) < 0.0000000001)) break : missing value where TRUE/FALSE needed I was able to track the error to the call to optim within normexp.fit but that is as far as I got. Unfortunately, it is particular to this array, as for other datasets on the same platform it works fine. Not sure how to attach a 200mb file to this post. I am working with an agilent 2 color array for chip-on-chip. The features were extracted with agilent's feature extraction software. Other background correction options work fine, including rma. The code: array <- read.maimages(files=file, source="agilent",columns= list( R = "gMeanSignal", G ="rMeanSignal", Rb ="gBGMedianSignal", Gb = "rBGMedianSignal"),other.columns=c("PositionX","PositionY")) array.norm <- backgroundCorrect(array,method="normexp") the output: Green channel Error in if (all(abs(delta) < 0.0000000001)) break : missing value where TRUE/FALSE needed Any idea? Many thanks, augusto -- Augusto Rendon PhD Research Associate ?Department of Haematology, University of Cambridge & National Health Services Blood and Transplant Cambridge Long Road Cambridge CB2 2PT Phone at EBI: +44 (0) 1223 49 4690 Phone at BSU: +44 (0) 1223 330 372