Hi all,
We've been discussing lately about if the enrichment score (ES), not the normalized (NES), can have values higher that 1 and lower than -1. I've done several tests using fgsea (R, package) and I never get an ES out of the range of the range of [1,-1].
Is it possible to get an ES outside this range? If this is possible, when can an ES be outside this range.
Thanks all for your time.
•
written
by
runningonempty •
1 answer
No, it's not possible. The maximal and minimal ES values are exactly 1 (when the selected genes are exactly the top up-regulated) and -1.
You can check out the section "4.1 Geometric interpretation of GSEA statistic" of fgsea preprint (https://www.biorxiv.org/content/10.1101/060012v1.full) for the better understanding on why it is so.
Log in to answer this question.
-
Comparing GSEA results across multiple cohorts
written by Picasa
-
Rescaling normalized enrichment score (NES)
written by CTLong •
-
RNA-seq GSEA cutoff
written by maplewj •
-
Gene Set Variation Analysis (GSVA)
written by Kyle •
-
Does Normalized Enrichment Score (NES) from GSEA correlate with gene set upregulation and down regu…
written by grsyhhsb •
-
How to reverse the color scale of a dot plot made by ggplot2?
written by vHelmholtz •
-
fgsea tutorial issue with BiocParallel
written by 1769mkc
-
Negative normalized Enrichment Score (NES) in GSEA analysis
written by Phil S.
-
question about GSEA
written by kanwarjag
-
The range of enrichment score and Normalized Enrichment Score in gene set enrichment analysis softw…
written by Zhilong Jia
