Hello, 

I performed an analysis of differential expression with SAM method, obtaining the genes that are up and down. I used these scripts:

> samfit = SAM(exprdc, group, resp.type="Two class unpaired", fdr.output=.01)

> sigSAM_low = as.numeric(samfit$siggenes.table$genes.lo[ ,2])

> sigSAM_up = as.numeric(samfit$siggenes.table$genes.up[ ,2])

> sigSAM = append(sigSAM_up, sigSAM_low)

> ID = featureNames(arraysdc.rma)[sigSAM]

>  ID[ID == "NA"] = NA

The next point was put the name of these genes, for it I saw the annotation and load the library:

"pd.hugene.2.0.st"

> source("http://bioconductor.org/biocLite.R")

> biocLite("pd.hugene.2.0.st")

> library(pd.hugene.2.0.st)

Annotation loaded, the next step was to name the genes, but:

> symbol = as.character(lookUp(ID, "pd.hugene.2.0.st", "SYMBOL"))

Error in mget(x, envir = getAnnMap(what, chip = data, load = load), ifnotfound = NA) : 
  error in evaluating the argument 'envir' in selecting a method for function 'mget': Error in (function (classes, fdef, mtable)  : 
  unable to find an inherited method for function ‘columns’ for signature ‘"AffyGenePDInfo"’

And I can't do to get the names of the genes... I think that the problem is the annotation "pd.hugene.2.0.st" but I don't know.

Sorry for the inconvenience, and thank very much for the help.

Santi.

The pd.hugene.2.0.st package isn't an annotation package - it's a pdInfoPackage, used by oligo to map probes to probesets. Assuming that you summarized your data at the core (or transcript) level, you want to annotate using the hugene20sttranscriptcluster.db package.

In addition, the lookUp() function is probably not what you want to use. If you simply want HUGO gene symbols, then you can use either mapIds() or select(). The difference between those two functions has to do with how duplicate mappings are handled (e.g., there are many probesets on that array that can either measure multiple things, or that measure just one thing that has multiple HUGO symbols). The default for mapIds() is to just give you the first symbol, but there are other things you can do - see the help page - and the default for select() is to just give you all the probeset -> symbol mappings, so there will be duplicates that you have to handle yourself.

If you just want one type of annotation, mapIds() is the way to go, but if you need to annotate to multiple things (Entrez ID, Ensembl ID, gene name, etc), then you have to use select(), and deal with the duplicates however you see fit.