Dear all, I tried to use the adaptive background correction for a set of 22 arrays (2 colour). For array no 17 I get the following error RGb<-backgroundCorrect(RG, method="normexp", offset=50) Error in fit.normexp(foreground = RG$R[, j], background = RG$Rb[, j]) : optim failure: degeneracy of the Nelder-Mead simplex There is nothing obviously wrong with that array so is it the array that's causing the problem or something else? And what is the Nelder-Mead simplex???? I am running R 2.0.1 + Bioconductor under Debian testing and limma 1.8.16 Thanks, Anja -- Anja E. Schiel, Ph.D. Departments of General Internal Medicine and Human Genetics Leiden University Medical Center PO Box 9503 2300 RA Leiden The Netherlands tel: -31-(0)71-5276067 fax: -31-(0)71-5276075

>Date: Mon, 04 Apr 2005 12:10:48 +0200 >From: Anja Schiel <a.e.schiel@lumc.nl> >Subject: [BioC] Adaptive background correction error >To: bioconductor@stat.math.ethz.ch > >Dear all, >I tried to use the adaptive background correction for a set of 22 arrays >(2 colour). For array no 17 I get the following error > >RGb<-backgroundCorrect(RG, method="normexp", offset=50) > >Error in fit.normexp(foreground = RG$R[, j], background = RG$Rb[, j]) : > optim failure: degeneracy of the Nelder-Mead simplex > >There is nothing obviously wrong with that array so is it the array >that's causing the problem or something else? Adaptive background correction requires an iterative algorithm to estimate the background, and this algorithm sometimes fails. It is unlikely to be a problem with your data. This problem has been discussed before on this mailing list -- search the archives for "normexp". Newer versions of limma use a different iterative algorithm to estimate the "normexp" background. Try upgrading and see if the new version works for you: install.packages("limma") >And what is the Nelder-Mead simplex???? The online help tells you that the Nelder-Mead simplex algorithm is being used for the optimization -- follow the link which says "for more details see fit.normexp". You do not need to know the details of the algorithm. Gordon >I am running R 2.0.1 + Bioconductor under Debian testing and limma >1.8.16 > >Thanks, Anja > > >-- >Anja E. Schiel, Ph.D. >Departments of General Internal Medicine and Human Genetics >Leiden University Medical Center >PO Box 9503 >2300 RA Leiden >The Netherlands >tel: -31-(0)71-5276067 >fax: -31-(0)71-5276075