Hi Wolfgang, Thank you for you for your help with the volcano plot, I was able to get it to work. I am trying to make Figures on variance estimation as shown in Supplementary II (4.3). What function should I use for 'varianceFitDiagnostics' ? Many Thanks, Melissa Dear Melissa does the code work for you if you reproduce it exactly from the example? Once that is established, you could start making changes and then see when what breaks. Also, why do you want to set log="x"? The data that you are supplying for the x-axis, log2FoldChange, are already on the log-scale, so that argument does not seem necessary (in fact: wrong). Also, if the complexity of lattice of plots is overwhelming, you can always use the basic R plots: plot( log2FoldChange, -log10( pval ) ) Best wishes Wolfgang

Hi Melissa the .Rnw file that generated Supplement II is contained within the tar-ball 'Additional file 3' on http://genomebiology.com/2010/11/10/R106/additional This was done with DESeq DESeq_1.1.12, and if your aim is to exactly reproduce that supplement, please use versions of all packages (and of R) matching the sessionInfo() output at the begin of that document. If your aim is data analysis, then please a more recent version of R, Bioconductor and DESeq. In more recent versions of DESeq, the function varianceFitDiagnostics has been rendered into an inactive stub, and the package vignette shows you newer and better suggestions on how to check fit quality. Best wishes Wolfgang. Mar/27/12 1:53 PM, Melissa.Martin at lshtm.ac.uk scripsit:: > Hi Wolfgang, > > Thank you for you for your help with the volcano plot, I was able to get it to work. > > I am trying to make Figures on variance estimation as shown in Supplementary II (4.3). > What function should I use for 'varianceFitDiagnostics' ? > > Many Thanks, > > Melissa > > > > Dear Melissa > > does the code work for you if you reproduce it exactly from the example? > Once that is established, you could start making changes and then see > when what breaks. > > Also, why do you want to set log="x"? The data that you are supplying > for the x-axis, log2FoldChange, are already on the log-scale, so that > argument does not seem necessary (in fact: wrong). > > Also, if the complexity of lattice of plots is overwhelming, you can > always use the basic R plots: > > plot( log2FoldChange, -log10( pval ) ) > > Best wishes > Wolfgang > > > > _______________________________________________ > 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 -- Best wishes Wolfgang Wolfgang Huber EMBL http://www.embl.de/research/units/genome_biology/huber