[BioC] PANTHER in R.... Gene lists in sigPathway?

Felipe Riveroll Aguirre friveroll at gmail.com
Tue May 22 18:45:19 CEST 2012


May be you can get the limma logFC associated values, for associated
GWAS genes, in the example from SPIA, I started only with entrezID's
from KEGG link, and grab the logFC values from limma in sig_genes
vector

sig_genes <- subset(top, adj.P.Val<0.01)$logFC
names(sig_genes) <- subset(top, adj.P.Val<0.01)$ENTREZ

with this code

genes.both <- sig_genes[which(names(sig_genes) %in% genes.via.sel)]

Where  genes.via.sel, was a gene list set of entrezID's to get limma
associated values.

So you can do an approach like this to get associated logFC values
from associated GWAS genes

Also there is another server I known for gene lists

http://www.bioprofiling.de/R_spider.html

Here you get an analysis combining signaling and metabolic pathways
from Reactome and KEGG databases.

So, R only helps to get the required inputs, for external web appz, or
you can download pantherdb database and write an script for automate
the whole process, from R output

Best,

Felipe

2012/5/22 Ovokeraye Achinike-Oduaran <ovokeraye at gmail.com>:
>
> Hi Felipe,
>
> Thanks. The thing is that I have genes from both GWAS and DE analysis from Limma in one list. So they don't all have logFC values, hence my question about gene lists and not necessarily gene sets. I ran my list through the PANTHER web interface and got some results but I just wanted to know if it had an R implementation.
>
> DAVIDQuery() might work if it gives the same options as is web interface of choosing the PANTHER pathway option but I didn't find the instruction manual very straightforward.
>
> Thanks again.
>
> -Avoks
>



More information about the Bioconductor mailing list