[BioC] problem with hgu133a // GOstats // BioC-R2.1.1

Wolfgang Huber huber at ebi.ac.uk
Tue Aug 2 15:35:48 CEST 2005


Hi Katleen,

For question 1, try on www.bioconductor.org. Also, try the new 
"biocLite" script that automagically finds most packages.
http://www.bioconductor.org/ -> Install How To

For question 2, we all know that genome annotations, Unigene etc. change 
over time, and the three different responses you get seem to reflect 
that. Have a look at the package documentation of hgu133a, i.e. type:
   "? hgu133a"
for details. In particular the data in the hgu133a package is based on 
EntrezGene, which might explain the discrepancy with Ensembl. The people 
that build these packages simply take what the public genome databases 
provide, we have no resources for correcting their mistakes or in fact 
for any kind of manual curation.

Still, it is of course annoying if you care especially for that gene - 
in that case there is no other way than curating (that part of) the 
metadata yourself, perhaps using the AnnBuilder package, or the biomaRt 
package.

For question 3, I pass (don't use MS-Window), but the above-mentioned 
script biocLite might do.

  Cheers
   Wolfgang


> 
> *question 1: GOstats*
> When I try to use library(GOstats) in R2.1.1 (with recent installation 
> of BioC: getBioC()), I get following message:
> 
> Loading required package: graph
> Error: package 'graph' could not be loaded
> In addition: Warning message:
> there is no package called 'graph' in: library(pkg, character.only = 
> TRUE, logical = TRUE, lib.loc = lib.loc)
> 
> Where can I find the graph package (I haven't found it on the CRAN website)?
> 
> *question 2: hgu133a*
> Is it possible that the hgu133a package is not up to date: for example 
> when I give the command aafSymbol("213268_at","hgu133a") following 
> results appear:
> /*PER3 */when I use R 2.1.1 (with recent installation of BioC: 
> getBioC()) and */UTS2/* when I use R 2.0.1 (with recent installation of 
> BioC: getBioC("all")). However, on the Ensembl website 213268_at = 
> */CAMTA1/*.
> 
> Another error message appears with R 2.0.1 when I give the command: 
>  >aafSymbol("213268_at","hgu133a"):
> 
> Loading required package: reposTools
> [1] "_You have package hgu133a but the incorrect version_"
> An object of class "aafList"
> [[1]]
> An object of class "aafSymbol"
> [1] "UTS2"
> 
> *question 3: BioC-R2.1.1*
> The commands Install or  Update packages from Bioconductor  under the 
> Packages menu-command-bar is not available anymore in the R2.1.1 version 
> compared to the R2.0.1 version? Is it normal or did I forgot something 
> when I installed BioC or R?
> 
> Thank you in advance,
> Best regards,
> Katleen
> 


-- 
Best regards
   Wolfgang

-------------------------------------
Wolfgang Huber
European Bioinformatics Institute
European Molecular Biology Laboratory
Cambridge CB10 1SD
England
Phone: +44 1223 494642
Fax:   +44 1223 494486
Http:  www.ebi.ac.uk/huber



More information about the Bioconductor mailing list