Entering edit mode
Hi all,
fgseaRes <- fgsea(pathways = c2_list, stats = stats, minSize=15, maxSize=500)
significantResults <- fgseaRes[fgseaRes$padj < 0.25, ]
The significantResults changes everytime I run this chunk. Maybe because of permutation. So the significant pathways change, so I can't say which pathways are significant. Would you please have a suggestion? Thank you so much!
Thanks ATpoint! Setting a fixed seed can reproduce the same result. However, if someone else analyze the same data with different seed, their result is inconsistent with my result. I plan to further narrow down pathways from significant pathways so I set padj < 0.25. From Gordon answers, seem p value from fgsea is not significant as it is. I tried the function he suggested: camera(). The pathways I got were different with fgsea. I would like to find which pathways are different between 2 conditions so do you think camera() is the best choice? We also have other method for pathway analysis like gene set variation analysis.