If the interest is really at *normalization*: save the target distribution estimated from the training set, then use preprocessCore::normalize.quantiles.use.target. b On 13 January 2012 14:08, James W. MacDonald <jmacdon at="" med.umich.edu=""> wrote: > Hi Daniel, > > > On 1/13/2012 8:47 AM, Daniel Brewer wrote: >> >> Hello, >> >> We have done some analysis on a set of Affy microarray data normalised by >> RMA and produced a predictor. ?We would like to test this predictor on a >> training set we have. ?Is it possible to RMA normalise the test dataset so >> that the probes have the same distribution as the training dataset without >> normalising all the data together? ?Our concern is that if you normalise >> them all together then this would mean we would have to go through all the >> analyses again. > > > If you had used the frma package for the initial processing, then yes. There > may even be some way to use frma with your extant processed data, but you > would have to look to see. > > Best, > > Jim > > >> >> Thanks >> >> Dan >> > > -- > James W. MacDonald, M.S. > Biostatistician > Douglas Lab > University of Michigan > Department of Human Genetics > 5912 Buhl > 1241 E. Catherine St. > Ann Arbor MI 48109-5618 > 734-615-7826 > > ********************************************************** > Electronic Mail is not secure, may not be read every day, and should not be > used for urgent or sensitive issues > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor