[BioC] Rredland installation difficulties

Paul Shannon pshannon at systemsbiology.org
Thu Apr 30 18:34:13 CEST 2009


Vince and Patrick,

Thanks to you both for your replies.

I propose to install Rredland on a linux box, and try out the idea of  
using it to obtain networks from NCI.  That ought to work, right? --  
Rredland is known to work when built from source on linux?

If NCI proves to be a good source for cancer-related pathways, then  
maybe that would justify including mac.binary.leopard builds at some  
time in the future.

  - Paul

On Apr 29, 2009, at 7:42 PM, Vincent Carey wrote:

> for what it's worth, librdf has supported sqlite stores for some  
> time.  we might consider dropping
> the requirement for the BDB back end and use sqlite henceforth.  if  
> i knew there were users
> i would explore this; a caveat is that the redland examples of the  
> sqlite interface were not
> very clear when i last checked.
>
> On Wed, Apr 29, 2009 at 9:25 PM, Patrick Aboyoun  
> <paboyoun at fhcrc.org> wrote:
> Paul,
> We currently don't support Rredland in a mac.binary.leopard format  
> because we don't have the infrastructure in place to support it on  
> the build machine. Since we build universal binaries (i386, x86_64,  
> ppc) on Mac OS X Leopard we usually have to build 3rd party  
> libraries ourselves, unless generous people in the user community  
> build them for us. (Thanks Gtk+ team!) With Rredland we need to  
> install both redland and Berkeley DB on our build machine. I have  
> redland in place, but need to do some research getting Berkeley DB  
> installed. If we do move to support Rredland in a mac.leopard.binary  
> form, it will more than likely be with custom builds with libraries  
> located in /usr/local, not /opt/local as darwin ports likes to use,  
> just as the Tiger build is structured. Have you tried to install the  
> package from source? If so what results have you had?
>
> Thanks for the tip on the misdirected link. I suspect the Reference  
> Manual wasn't built properly and this the result. I'll look to  
> correct this.
>
>
> Patrick
>
>
>
> Quoting Paul Shannon <pshannon at systemsbiology.org>:
>
> I wish to read BioPAX level 2 pathway files from the National Cancer
> Institute's 'Pathway Interaction Database'.  Rredland was recommended
> to me which looks looks very promising.  Perhaps these pathways can
> even be expressed as a bioc graph...
>
> I have enountered some of problems, leading to these questions and  
> comments:
>
> 1) Lacking librdf, I used darwin ports to install it on my MacBook
> (Leopard) laptop.  This installed lots of libraries to /opt/local/lib.
> Rredland seems to want dynamic libraries in /usr/local/lib.  Is this a
> mac os x tiger/leopard difference?   Do I run into this because I used
> the Tiger binary, in the absence of a Leopard binary?  Is a Leopard
> binary planned for the future?  Shall I try to build from source?
>
> 2) (minor) The 'Reference Manual' link at
> http://www.bioconductor.org/packages/release/bioc/html/Rredland.html
> points to       http://www.bioconductor.org/packages/release/bioc/,
> rather than to a document.
>
> Thanks!
>
>  - Paul
>
> _______________________________________________
> 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
>
> _______________________________________________
> 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
>
>
>
> -- 
> Vincent Carey, PhD
> Biostatistics, Channing Lab
> 617 525 2265



More information about the Bioconductor mailing list