Dear all,

I am so sorry that I'm almost reposting my question. I performed normalization of an HTA 2.0 array and I wanted to perform non-specific genefiltering. But I met an error which I don't know how to deal with.

My codes are:

>library(pd.hta.2.0)

>library(oligo)

>GSE12345cels = list.celfiles()

>GSE12345raw = read.celfiles(GSE12345cels)

>GSE12345.rma = rma(GSE12345raw, target='core')

>GSE12345exp<-exprs(GSE12345.rma)

>library(affycoretools)

>GSE12345.rma<-annotateEset(GSE12345.rma,pd.hta.2.0,type = "core")

>GSE12345pData=read.csv("/Users/lian/Documents/GSE12345/GSE12345pData.csv",row.names=1, header=T)

>metadata <- data.frame(labelDescription=c("Disease state","Tissue origin"),rownames=c("AFSR","Tissue"))

>GSE12345phenoData <- new("AnnotatedDataFrame",data= GSE12345pData, varMetadata=metadata)

>GSE12345annotation <- "pd.hta.2.0"

>GSE12345expressionSet <- ExpressionSet(assayData= GSE12345exp,phenoData= GSE12345phenoData, annotation= GSE12345annotation)

>library(genefilter)

>GSE12345f <- nsFilter(GSE12345expressionSet)

And here is the error:

Error in (function (classes, fdef, mtable)  :
  unable to find an inherited method for function ‘columns’ for signature ‘"AffyHTAPDInfo"’

Could anyone please give me some suggestions?

Thank you very much.

The annotation slot for an ExpressionSet is supposed to contain the name of a package that can be used to map probeset IDs to information about the genes/transcripts the probesets are intended to measure. For this array that would be the hta20transcriptcluster.db package (for the current release I incorrectly called this package the hta20sttranscriptcluster.db package - note the extra bolded ST in that name - so you need to use the correct package name, dependent on the version of BioC you are using).

The pd.hta.2.0 package is used to map probes to probesets when you are summarizing the data, and isn't the kind of package that nsFilter is looking for.