gene-to-GO mappings with topGO gene set enrichment analysis

Question

[jose.wo]

Hello all,

I've been using the topGO package from the bioconductor to do GSEA. Today I compared the results of topGO with those using panther with defaults. I general they are more or less the same but some of the GOTerms from panther GSEA are more meaningful for my analysis. I could use panther but I would prefer to use R/Bioconductor so I can pipeline my analysis more easily.

This is the code I used in R:

selection <- function(allScore){return(allScore < 0.05)}

GOdata <- new("topGOdata",
               ontology      = "BP",
               allGenes      = IdentifiedGenesVector, 
               geneSel       = selection,
               nodeSize      = 5,
               annotationFun = annFUN.org,
               mapping       = "org.Hs.eg.db",
               ID            = "symbol")

And this are the first few genes I want to analyze:

head(IdentifiedGenesVector)
RP11-27K13.3   AC021224.1        ALCAM      ALDH1A1        DHRS9        IFI27
         0.04         0.04         0.04         0.04         0.04         0.04

Can I use panther's mapping in topGO?

Is there any other gene-to-GO mapping that I could try?

Is there a generally accepted mapping to use?

I have seen in many posts that many people cast their own mapping list but I'm not sure how I would do this.

Thanks a lot, Jose.

Answer 1

[Kevin Blighe]

Hola / Olá Jose,

There are many GO enrichment programs to use, with no standard existing for which one to use. There is neither any program that is better than the other. topGO is one of the most popular, though.

If you wanted to use PANTHER, then you can simply use panther.db package, also in Bioconductor:

• http://bioconductor.org/packages/release/data/annotation/vignettes/PANTHER.db/inst/doc/PANTHER.db.html

Kevin