Hi, I look at the slot genes in fit2 (fit2$genes) and selected.dif. Both have the some number of genes. But, when I try fit2$genes[selected.dif,] I get a error message: Error in "[.default"(fit2$genes, , selected.dif) : incorrect number of dimensions If I use topTable and set the coef slot to 1, 2 or 3 I get the DE genes, but, I suppose that in these case the list of genes are different from the example in usersguide. Or not? I make 3 contrast before fit the values, then I get 3 coef in fit2. Could be this the cause of error message? In the usersguide there are 2 contrasts. Thanks Marcelo Marcus Davy escreveu: >The gene annotation information is usually a list component called 'genes' within the output object from lmFit, and eBayes. This gets used in the construction of the table of the top-ranked genes using topTable. > >Have a look at fit2$genes (for your object fit2). > >Your R expression has just provided a logical vector of corrected p-values for genes <0.05 based on the B&H's stepup FDR control, try: > >fit2$genes[selected.dif,] > > >Marcus > > > >>>>Marcelo Luiz de Laia <mlaia@fcav.unesp.br> 17/02/2005 12:08:53 PM >>> >>>> >>>> >Hi, > >I have been analyse a time course experiment with limma and I get the >end of section 16 in userguide very well. > >But, I don't know how do I do to get the names of genes differentially >expressed! > >For example, I do: > >selected.dif <- p.adjust(fit2$F.p.value, method="fdr") < 0.05 > >In object *selected.dif* I see that there are a lot of TRUE, but I don't >know is their names! > >Any help is very appreciated. > >Marcelo > > > > -- No virus found in this outgoing message. Checked by AVG Anti-Virus.