Making geneLists for gseKEGG()/gseGO()/GSEA()

[Lil-Psilocybe]

Hello! Thanks for a fantastic package! I am having a lot of confusion about the input gene and annotation term info though. I cross-posted this in Biostars , but would like some advice from the folks who wrote the program.

Basically my question boils down to this: for any annotation used (GO/KEGG/etc.), does each GO/KEGG number identifier (GO:5478201, K12345, etc.) have to be associated with a single gene? I'm generating my annotation info (for my non-model organism) from EggNOG mapper, which spits out multiple K#s and GO#s for each queried entry like so:

I got gseKEGG() to run when I reduced the amount of K#s-per-gene like so:

but am having trouble getting gseGO()/GSEA() to run since I still have multiple GOIDs for each gene like so:

For GSEA(), TERM2NAME looks like this:

And TERM2GENE looks like this:

With all this in mind, do I need to reduce my GO numbers to just a single GO number for each single gene? And if so, how do I account for the GSEA statstistic being tested since this is specific to each gene?

Thanks for any recommendations!