ClusterProfiler GeneRatio demoninator not the same for all values

Question

[Lily]

Hello,

I was running clusterprofiler in order to see enriched pathways/modules using the KEGG database. One thing I noticed after running everything and looking at my final data output with the GeneRatio and BgRatio is that the denominator of my GeneRatio is not always the same for all pathways/modules.

For example:

Menaquinone biosynthesis, modified futalosine pathway   **25/7645** 251/198201
Multidrug resistance, efflux pump MexJK-OprM    **25/7645** 282/198201
C10-C20 isoprenoid biosynthesis, plants **31/7645** 412/198201
Photosystem II  **108/4163**    326/198201
Rebeccamycin biosynthesis, tryptophan => rebeccamycin   **48/4163** 146/198201
Pyrrolnitrin biosynthesis, tryptophan => pyrrolnitrin   **48/4163** 147/198201

Where the first ratio is the GeneRatio and the second ratio is the BgRatio. The BgRatio all have a denominator of 198201 which is to be expected. But my denominator changes for the GeneRatio (ex some are 7645, some are 4163...)

Anyone know why this is occurring and how to fix it? Thanks!

[mark.ziemann]

That sure sounds strange. Could you give the fora() function a try and see if it gives the expected results? link: https://rdrr.io/bioc/fgsea/man/fora.html

[Lily]

I ended up realizing what had happened. I was following a tutorial where I created two lists (one for up and one for down regulated genes) in one object and ran together and so when I ran clusterProfiler, I am assuming that the GeneRatio is based off the number of either up or down regulated genes. But, since my background gene set contained everything, the BgRatio was fine.