Hi, all, I'd like to background correct my data using sequence info and then to explore them with different normalization algorithms, vsn and quantiles, for examples. Has anyone ever tried it this way? Can someone give some suggestions of how to do that? Thank you in advance for your help. I performed background correction of my data using "bg.correct.gcrma" function and used "expresso" function to normalize (quantiles) the data without background correction. To my surprise, the normalized intensities from this procedure are quite different from the intensities generated with gcrma function (default parameters). Using separate bg correction seems to generate many extremely small intensities (~ -9 of log2 value). Is there anything wrong of what I have done? Best, Puhong Gao, PhD Dev. Psychobiology NYSPI, Unit# 40 1051 Riverside Dr. New York, NY 10032 (212) 543-5710 [[alternative HTML version deleted]]