Should gene2cat have all the assayed gene IDs?

Question

[Kazuna]

Hi, I have a question about goseq.

I am now analyzing the RNA-seq data of a non-supported organism. I learned from goseq manual (goseq: Gene Ontology testing for RNA-seq datasets) that a data frame of GOs has to be provided in my case. Since I just have a transcriptome fasta without annotation, I used Blast2GO to annotate DEGs. My question is, do I have to annotate all the assayed genes or is it okay to just annotate DEGs?

GOTestWithCat <- goseq(pwf, gene2cat = gene2cat, use_genes_without_cat=TRUE)

This gene2cat only contains DEGs. When I ran this code, I got numDEInCat * 100 / numInCat = 100 for every categories. Now I'm guessing I needed to provide GO IDs for every assayed genes and not only for DEGs. Is this right?

It would be great if someone would answer my question. Thank you in advance!