Dear all, My goal is to obtain the standard errors of the signals given by RMA and GCRMA. - for RMA, I used the function expresso [background correction= rma, normalization= quantiles, PM correction= PMonly, expression= medianpolish ] and then the se.exprs() one. I obtained a matrix with coefficients for each genes on each array. My first question is: are these coefficients in log2 base or not ? - for GCRMA, I did the same, with the command gcrma() instead of expresso() of course, and it seems that I cannot obtain the SE coefficients (I have "NA" everywhere). So my second question is: is it possible to obtain SE of the signals given by GCRMA and, if yes, how ? Thanks a lot, Florence.

I had the same problem with GCRMA, and this remind me another, more general, question... It's possible (and how it is possible) to control GCRMA procedure, and to have the different outputs (as in the other methods)? I wasn't able to find answers in the documentation, and on the web too, for the present version of GCRMA. Thanks in advance, Davide ----- Original Message ----- From: "Florence Combes" <florence.combes@paris7.jussieu.fr> To: <bioconductor@stat.math.ethz.ch> Sent: Monday, September 20, 2004 3:13 PM Subject: [BioC] SE in RMA and GCRMA > Dear all, > > My goal is to obtain the standard errors of the signals given by RMA and > GCRMA. > > - for RMA, I used > the function expresso [background correction= rma, normalization= > quantiles, PM correction= PMonly, expression= medianpolish ] and then > the se.exprs() one. > I obtained a matrix with coefficients for each genes on each array. > > My first question is: are these coefficients in log2 base or not ? > > - for GCRMA, I did the same, with the command gcrma() instead of > expresso() of course, and it seems that I cannot obtain the SE > coefficients (I have "NA" everywhere). > > So my second question is: is it possible to obtain SE of the signals given > by GCRMA and, if yes, how ? > > Thanks a lot, > > Florence. > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor >

the SE frome expresso is not a good choice. One can use affyPLM to get SE. For rma you can use fitPLM. It is a little complicated with gcrma right now that you need to background correct with gcrma, then use fitPLM with background=FALSE On Mon, 20 Sep 2004, Florence Combes wrote: > Dear all, > > My goal is to obtain the standard errors of the signals given by RMA and > GCRMA. > > - for RMA, I used > the function expresso [background correction= rma, normalization= > quantiles, PM correction= PMonly, expression= medianpolish ] and then > the se.exprs() one. > I obtained a matrix with coefficients for each genes on each array. > > My first question is: are these coefficients in log2 base or not ? > > - for GCRMA, I did the same, with the command gcrma() instead of expresso() > of course, and it seems that I cannot obtain the SE coefficients (I have > "NA" everywhere). > > So my second question is: is it possible to obtain SE of the signals given > by GCRMA and, if yes, how ? > > Thanks a lot, > > Florence. > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor >