Hello everyone! I've been doing some Differential Expression analysis on specific samples. It happens that I found a lot of genes that are DE. In total, of 24000 features, 11000 were up or down regulated in control vs group. Even tough the number is very high, I've very good reasons and evidences to believe these numbers.
My question is: how well ORA and GSEA algorithms handle large numbers of genes? I tried using cluster profiler with wikipathways, reactome, cell type and KEGG, but I only got results with p-values below 0.05 for KEGG with ORA analysis.
But, seing that half the genes are DE, I expected way more significative results, which made me question how well those methodologies handle very large number of genes!
Can anyone illuminate me?