Getting pval=0 for various enriched pathways after running GSEA

saezlab / decoupler-py

Python package to perform enrichment analysis from omics data.

https://decoupler-py.readthedocs.io/

GNU General Public License v3.0

157 stars 23 forks source link

Getting pval=0 for various enriched pathways after running GSEA #124

Closed saksham132 closed 4 months ago

saksham132 commented 4 months ago

Hi, I am running GSEA analysis for some single cell data. For many enriched pathways the pval = 0. The issue is I am not able to plot the result because of this 0 value, as it throws the runtime error due to (log(0)) in the calculation. I'm uncertain whether these p-values are genuinely 0 or if they represent extremely small values rounded to 0. Additionally, I might have made some error in the analysis earlier. Could you provide guidance on how to address this issue or suggest alternative plotting methods for such data? Thanks for any assistance!

code snippet:

PauBadiaM commented 4 months ago

Hi @saksham132 ,

Whenever you find p-values that are 0 in gsea it means that none of the randomly permuted values are higher than the original statistic. To try to solve this, you can:

increase the number of permutations (though increasing computing time)
set them to the minimum non-zero p-value in your results
use another enrichment method for example ulm which will be able to provide p-values

Hope this is helpful!

saksham132 commented 4 months ago

Thank you for the suggestions! I will try implementing these.