Hello everyone, I would like to seek help with a question regarding GO and KEGG enrichment analysis.

When conducting GO and KEGG enrichment analyses, what would be the best choice for the background gene set?For instance, all the genes of this species in the GO database, or all the genes expressed in tissues, or all the genes expressed in tissues and also annotated in the GO database.

Thank you for your help.

It should imo be the genes that were eligable to yield a "significant" result in your analysis. So, if you test set is derived from differential analysis, e.g. RNA-seq, then it should all tested genes intersecting the annotations you use, e.g. KEGG, GO, REACTOME, etc. If you have "just a list of genes", maybe from literature then I would at least try to find a background that properly reflects the situation. For example, a literature search find that in a certain celltype genes x,y,z... are consistently mentioned across studies, then maybe try to find genes that are "expressed" in this celltype. Something like that.