[BioC] AnnBuilder: ABPkgBuilder gives missing GO and NAs in ACCNUM

[Marc Carlson]

Hi Chuen,

I have not been successful in replicating your bug.  But I HAVE been
successful in building the package that you were trying to make.  I will
mail it to you in a personal email.

Here is the code as I ran it in linux (using the same packages that you
used).

library(AnnBuilder)
mysrcUrls<-getSrcUrl("all","Homo sapiens")
ABPkgBuilder(baseName="/home/mcarlson/tmp/Chuen_Tan/baseName3.txt",
srcUrls=mysrcUrls,baseMapType="ll",pkgName="chip89",pkgPath="/home/mcarlson/tmp/Chuen_
Tan",organism="Homo sapiens", version="1.1.0",author=list(author="Chuen
Seng",maintainer="<chuenseng.tan at ki.se>"),fromWeb = TRUE)

In the meantime, you might want to make double sure that your
environment is set up correctly.  There is a lot of information out
there about how to build R packages in windows.  Perhaps this FAQ might
be helpful? 

http://c*r*an.*r*-p*r*oject.o*r*g/bin/*windows*/base/*r*w-FAQ.html#*Package*s <http://cran.r-project.org/bin/windows/base/rw-FAQ.html#Packages>


    Good Luck,

        Marc




> Hi, Chuen,
>
> The purpose of the GenBank-environment (i.e. ACCNUM) is to record the probset 
> to GenBank Accession mapping provided by users via the base mapping file. It 
> will contain nothing but NAs when the base mapping type is "ll". ABPkgBuilder 
> currently doesn't extract GenBank Accessions based on Entrez Gene ID. This is 
> due to design choices rather than technical changes as I understood. 
>
> The missing of GO-environments is a real problem. The perl complaints right 
> after loading GO in your outputs is abnormal. Not sure whether it relates to 
> the GO problem. I know ABPkgBuilder only utilizes the perl scripts when it 
> maps probset IDs to Entrez Gene IDs. Everything is pure R when it gets to GO 
> annotations. You have already given the mappings as input, and you got most of 
> the annotations (PATH, CHR, etc), so it seems the perl problem is not the 
> cause of the GO problem. Sorry that I don't have more useful information :(
>
> best
>
> nianhua
>
> _______________________________________________
> 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
>   
> ------------------------------------------------------------------------
>
>
>
>