[BioC] bioconductor on rocks cluster

[David Schruth]

We use R on our Rocks Cluster.

R was installed across our compute nodes by folding its RPM into a rocks 
distro compute node (re)-build 
http://www.rocksclusters.org/roll-documentation/base/5.3/customization-adding-packages.html

We install our *libraries*, however,  manually & one-at-a-time into 
/share/apps/ via the root user
(who's R_LIBS variable points to the R_LIBS_SITE which points to 
/share/apps/R/site_library  which is

mounted on all nodes

)

For bioconductor packages we use BiocLite.
Just log in as root, fire up R, and run:

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

It sounds like others have done the same:
https://lists.sdsc.edu/pipermail/npaci-rocks-discussion/2007-September/027232.html

I suppose that you could install a the entire Bioconductor RPM in the 
same way we installed R (using a global node rebuild)  but that's 
probably overkill.

Anybody else have other ideas?


David M. Schruth
Informatics Consultant
Center for Environmental Genomics
University of Washington



Bryan Hepworth wrote:
> Hi All
>
> I'm curious to know if anyone has done a bioconductor install onto a rocks cluster at all. I'd like to know your experience and a how-to if you have managed to do so.
>
> Thanks
>
> Bryan
>
> Bryan Hepworth
> Assistant Institute Computing Officer
> Institute of Human Genetics
> Newcastle University
> International Centre for Life
> Newcastle
> NE1 3BZ
>
> _______________________________________________
> 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
>