[BioC] AnnBuilder problem?

[Ting-Yuan Liu]

Hi Denise,

For the UCSC part: It is us to force the build of Golden Path (GP) to be 
NA in the BioC 1.7 release.  Before the 1.7 release, the build of GP is 
the url of the GP source data.  It will cause some problem when running 
the R CMD check on the annotation packages because there is an "_" in the 
url.  Moreover, it doesn't make sense to put the url again in the built 
slot because the url is already in the documentation and the build slot is 
designed for the version (or date, etc) of source data.  Before we can 
find a better information for the GP build slot, please ignore the warning 
message for a while.

For the KEGG part:  KEGG anonymous ftp site is too busy (more than 400 
users) to be accessed on Monday, October 31, PST.  I have the same problem 
on Monday, and I wrote an email to KEGG for the issue.  KEGG fixed the 
problem on Monday night (PST).  Please try again to see if you still have 
this kind of problem.

Sincerely,
Ting-Yuan
______________________________________
Ting-Yuan Liu
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
Seattle, WA, USA
______________________________________

On Mon, 31 Oct 2005, Denise Mauldin wrote:

> 
> Hey all,
> 
> I'm having a problem with Annbuilder:
> 
> > tempConfig <- tempfile()
> > writeSourceUrlConfig(tempConfig) 
> > root = "http://sioux.fhcrc.org/MIRROR"
> > urlData = readSourceUrlConfig(tempConfig, root)
> > origUrls = getOption("AnnBuilderSourceUrls")
> > geoUrl = origUrls[["GEO"]] 
> > urlData[["GEO"]] = geoUrl
> > options(AnnBuilderSourceUrls=urlData) 
> > localMirror <- paste(root, "ftp.arabidopsis.org/home/tair/", sep="/")
> >
> > ABPkgBuilder(baseName="mpedbadm_genbank.txt",
> + baseMapType="gbNRef",
> + pkgName="MPEDB",
> + organism="Mus musculus",
> + otherSrc=c(UG="mpedbadm_unigene.txt"),
> + pkgPath="./",
> + version="0.1",
> + author=list(authors="Denise Mauldin", maintainer="Denise Mauldin <dmauldin at fhcrc.org>"))
> 
> Warning message: 
> Built for UCSC is not valid! in: getUCSCBuilt(organism)
> Error in parseKEGGGenome() : Faild to obtain KEGG organism code
> 
> >sessionInfo()
> R version 2.2.0, 2005-10-06, sparc-sun-solaris2.9
>  
> attached base packages: [1] "tools"  "methods"  "stats"  "graphics"  
> "grDevices" "utils" [7] "datasets"  "base"
>  
> other attached packages: 
>    AnnBuilder annotate    XML       Biobase      GO
>    "1.8.0"    "1.8.0"   "0.97-8"    "1.8.0"   "1.10.0" 
> 
> 
> ----
> Warning issue with UCSC:
> 
> If you browse to the MIRROR directory, it exists and has the Mus_musculus 
> subdirectory.  
> 
> If I look at the code in 1.8.0 Annbuilder for getUCSCBuilt() it says:
> 
> getUCSCBuilt <- function(organism){
>     url <- getSrcUrl(src = "GP", organism = organism) 
> 
>     built <- NA
>  
>     if(is.na(built) || is.null(built) || built == ""){
>         warning("Built for UCSC is not valid!")
>         return("N/A") 
>     } else { 
>         return(built)
>     }
> } 
> 
> So it seems like this was changed to always be nothing and always throw 
> this error?  (The rest of the code has been commented out.)
> 
> --- 
> Error issue with KEGG
> 
> However, it doesn't die on that error, it dies on the following KEGG 
> error.  This seems to be because even though I altered the URLs, it's 
> using the original URLs (which shouldn't that work anyway?).  I then 
> altered my ABPkgBuilder call to include srcUrls = urlData (as that's how 
> the ABPkgBuilder documentation now reads), but this throws a separate 
> error:
> 
> Error in validObject(.Object) : invalid class "GO" object: invalid object 
> for slot "srcUrl" in class "GO": got class "list", should be or extend 
> class "character"
> 
> > mode(urlData[["GO"]])
> [1] "character"
> > class(urlData[["GO"]])
> [1] "character"
> > urlData[["GO"]]
> [1] "http://sioux.fhcrc.org/MIRROR archive.godatabase.org/latest"
> 
> So this seems to be something in the code in one of the KEGG calls?  
> 
> Anyone have any ideas on what I'm doing wrong?  Anyone have any ideas on 
> what to do next?  
> 
> Thanks,
> Denise
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
>