[BioC] SPIA, hacked to run with only selected pathways

Tarca, Adi atarca at med.wayne.edu
Thu Dec 10 17:23:02 CET 2009


-----Original Message-----
>From: Paul Shannon [mailto:pshannon at systemsbiology.org] 
>Sent: Wednesday, December 09, 2009 5:36 PM
>To: Tarca, Adi
>Subject: SPIA, hacked to run with only selected pathways

>I have been using the bioc SPIA package, which I find very useful.
>After running a few sets of genes, and seeing their impact on all 83 KEGG pathways used by SPIA by default, I wanted to repeat my analyses at different 
>thresholds for differential expression, but only on a few pathways.  The 'spia' function takes quite a long time to run on my laptop.
>In order to do this, I hacked the spia function so that it would use a custom, smaller 'path.info' with, in my case, just the 5 KEGG pathways that the 
>full,  earlier analyses had turned up.
>Is this hack of interest to you?
>Thanks for the very useful tool!
> - Paul Shannon
>   Institute for Systems Biology
>   Seattle


Dear Paul,

Thanks for the suggestion to allow SPIA to run only on selected pathways for the users that need this feature. I have added an optional argument in the spia function call (called "pathids") so that a subset of preselected pathways can be specified.

For e.g. if you want to analyse only the pathways with KEGG ids "05210" and "04010" using the colorectal cancer dataset from SPIA, use:

> library(SPIA)
> data(colorectalcancer)
> res<-spia(de=DE_Colorectal, all=ALL_Colorectal, organism="hsa",pathids=c("05210","04010"),nB=2000)
> res

It will take a day or two before the change in the SPIA package will be reflected in the bioconductor's devel repository.

Regards,
Adi 


 



More information about the Bioconductor mailing list