[BioC] Asking again: AnnotationDbi and GOstats - feed probe names to hyperGTest?

[Robert Gentleman]

Hi,
   I suspect that the issue is one of what IDs you are supplying and 
what it is expecting.  The error message points to,
getUniverseHelper
and you can see the source code from within R, via

Category:::getUniverseHelper

and you can do something like:
debug(Category:::getUniverseHelper)

to have execution stop when you enter this function. At that point, you 
can query to see what probes is, what datPkg is and what the entrezIDs are.

  I would then check to see what the value of univ is (the first 
command), and see how that compares with the supplied entrezIds - my 
guess is that these are not using the same IDs, and hence there are no 
matches :-(, but that might give you a hint as to how to fix things,
for your case,

best wishes
   Robert

ps things have changed a bit, and will change more soon, in devel, but I 
doubt those changes will relate to your problem - but as you give no 
version information..

Johannes Graumann wrote:
> I asked this a while ago, but didn't get any further ...
> 
> I built an annotation package for a mouse IPI database using "AnnotationDbi"
> (IPI.MOUSE.3.37.20080509.db) and am now putting it to use with "hyperGTests"
> from "GOstats".
> The annotation package mapps all IPI IDs as "probes" to the relevant
> annotation data and I thought it should be possible to hand "geneIds" and
> universeGeneIds in the form of probe names (here IPI IDs) to the hyperGTest
> function. If I do so I always get the error: 
> # Error in getUniverseHelper(probes, datPkg, entrezIds) : 
> #  No Entrez Gene ids left in universe
> If I use the annotation package to pre-convert the probes (IPI IDs) into
> entrez-IDs everything's just fine ... Am I making a mistake or does this
> not work inherently?
> 
> Thanks, Joh
> 
> params <- new(
>         "GOHyperGParams", 
>         geneIds = BunchOfIPIIds,
>         universeGeneIds = BiggerBunchOfIPIIds,
>         annotation = IPI.MOUSE.3.37.20080509.db,
>         ontology = "BP",
>         pvalueCutoff = 1.0,
>         testDirection = "over"
> )
> HGT <- hyperGTest(params)
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
> 

-- 
Robert Gentleman, PhD
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M2-B876
PO Box 19024
Seattle, Washington 98109-1024
206-667-7700
rgentlem at fhcrc.org