custom gene universe in gseGO and gseKegg functions

[nikofleischer]

Dear developers,

I love the clusterprofiler package and it simplifies my workflow for downstream analysis of clusters from single-cell data tremendously. I mainly use the gseGO and gseKEGG functions as I like that they include the FC and not only rely on ranked lists as enrichGO does. However, I am wondering why I can't set a gene universe in these functions and couldn't find further info in the manual. Do they consider only the genes present in the gene lists (so differentially expressed genes for each cluster as outputted by seurat) or all the genes present in the organism specified? I would like the former behavior, as the output that I got tends to be very broad and unspecific so far. Therefore it would be great if a feature to specify the gene background could be implemented in the gsea functions as well.

Thanks a lot for your help, best Niko

[lovebaboon1989]

You could refer to this link: https://yulab-smu.top/biomedical-knowledge-mining-book/universal-api.html?q=GSEA#msigdb-gsea the function enricher() and GSEA() enables you to perform an universal functional analysis on your own background gene sets.