How should I interpret a volcano-like plot from GSEA analysis (Functional Analysis module)

[eterlova]

Hello all, I am using a GSEA approach to run pathway analysis on my metabolomics data. this analysis produces the plot that reminds me a volcano plot with pathways arranged according to their p-value on the y-axis and NES on the x-axis. On this plot NES (i.e. enrichment scores) can be negative or positive. How is this determined? How should I interpret this plot?

Best,
Lisa
ZA17_DH_GESA.pdf (111.2 KB)

[jeff.xia]

Indeed, we use volcano plot as a visual summary of the GSEA result (based on its p values and NES). I guess your question is on how to interpret NES? I would refer you to the GSEA website. Here genes are replaced by putative metabolites

[EVC]

Dear Jeff

The GSEA website explains that the NES is computed as a ratio of experimental Enrichment Score and mean of ESs against all permutations

I am unclear on how a NES could be a negative value, as both numerator and denominator of that ratio are both positive values?

I found no example on the GSEA website of negative ES or NES. I probably missed something?

Regards

[jeff.xia]

Enrichment score can be positive or negative. Positive scores mean the pathway hits tend to be on the top half of the ranked list, while negative means the hits end to be at the bottom half. Note negative is not meaningful (interpretable) if you use p-values for ranking

Note we don’t recommend GSEA - it is well-established but lacks power - see our publication “Comprehensive investigation of pathway enrichment methods for functional interpretation of LC–MS global metabolomics data”