How to use topGO tool for enrichment analysis of non-model organisms

Question

[otieno43]

Hello guys, This may be the most simple thing but let me just ask.

There is something I seem not to get in the topGO manual. I have two files that I believe I need for topGO analysis. I am new in using topGO

1. A list of deferentially expressed genes with their associated p-values (from desq2).
2. I have generated customized annotation for every gene of my model. List of 6 $ GFUI18_000001: chr [1:10] "GO:0110165" "GO:0016020" "GO:0005743" "GO:0005739" ... $ GFUI18_000002: chr [1:5] "GO:0005634" "GO:0003723" "GO:0046872" "GO:0003676" ... $ GFUI18_000003: chr [1:17] "GO:0005829" "GO:0005524" "GO:0003779" "GO:0035091" ... $ GFUI18_000004: chr [1:13] "GO:0031083" "GO:0030133" "GO:0000139" "GO:0005509" ... $ GFUI18_000005: chr [1:44] "GO:0016020" "GO:0016021" "GO:0001664" "GO:0002177" ... $ GFUI18_000006: chr [1:23] "GO:0016020" "GO:0016021" "GO:0005347" "GO:0022857" ...

I seem not to understand how to use these two file to generate topGOData object and subsequent enriched GO terms for deferentially expressed genes.

some help please (sample script).

Answer 1

[dthorbur]

Unless things have changed since I last used topGO, you need an org.db package for your species that provides mapping between transcripts and their GO annotations.

So, if you must use topGO, you could try and make one of these packages, or use orthologs between your species and one of the model organisms that have an org.db package.

The alternative is to use a different tool. I've used gProfiler2 to conduct enrichment analyses in non-model organisms in the past. Here is the vignette about how to use it.

Also, if you want help scripting, provide a script that you've tried to do an analysis with, and show where you're getting stuck. Asking people to write you scripts without showing you've tried isn't going to get you very far.

Answer 2

[wdlin]

Just reached this page by searching TopGO. Here is the steps I will take, in case you are attempting to use TopGO elim method:

ALERT: potential self advertisements

1. translate the GO annotation data of all genes (from models) into a GOBU tree, as described in this wiki page I wrote.
2. Use GOBU and its FastInput plugin to enter the gene list. The MultiView plugin is for doing GO enrichment, including an implementation of the TopGO elim method.

Also, the wiki page includes command line method of doing GO enrichment.