Howdy, I am trying to create a subset of probe intensities from an AffyBatch object. As a new user to the affy package (and Bioconductor, in general) I was pleased to find an article presenting the package by Gautier et al. in Bioinformatics (2004) 20:307-315. I was *especially* pleased to find the R code in Fig 4. which seems to do exactly what I'd like: ## celdata is an AffyBatch of 75 arrays... controls.gni<-grep("AFFX.+", geneNames(celdata)) controls.names<-geneNames(celdata)[controls.gni] controls.indices<- indexProbes(celdata, which="pm", controls.names) ## so far so good....but: controls.intensities<-intensity(celdata)[controls.indices, ] gives: Error: invalid subscript ## ???? Am I missing something obvious? BTW: Inspection of control.intensities show that this is a list object. By unlisting, such as: x<-unlist(controls.indices) I can recover the data but they are in a (rather nasty) vector. thanks in advance, LL -- _____________________________________ Larry LaPointe PhD Student CSIRO / Flinders Univ / Enterix Inc. Molecular Sciences/Bioinformatics p (612) 9325 3250 m 04 0099 5027

Larry La Pointe wrote: > Howdy, > > I am trying to create a subset of probe intensities from an AffyBatch > object. As a new user to the affy package (and Bioconductor, in > general) I was pleased to find an article presenting the package by > Gautier et al. in Bioinformatics (2004) 20:307-315. I was *especially* > pleased to find the R code in Fig 4. which seems to do exactly what I'd > like: > > ## celdata is an AffyBatch of 75 arrays... > controls.gni<-grep("AFFX.+", geneNames(celdata)) > controls.names<-geneNames(celdata)[controls.gni] > controls.indices<- indexProbes(celdata, which="pm", controls.names) > > ## so far so good....but: arg.... there is a 'but'... > controls.intensities<-intensity(celdata)[controls.indices, ] > > gives: > Error: invalid subscript ## ???? > > Am I missing something obvious? Since they the time the paper was written, the API of the package changed slightly... do: controls.indices <- unlist(control.indices) right after the line controls.indices<- indexProbes(celdata, which="pm", controls.names) > > BTW: Inspection of control.intensities show that this is a list object. > By unlisting, such as: > x<-unlist(controls.indices) Yes (see above) > I can recover the data but they are in a (rather nasty) vector. "recover" ?? "nasty" ??? what do yo mean ? > thanks in advance, > LL > > L.