Ideally, when hypeR is called with the KS test, the hypeR-dot plot should color-code the dots based on whether a geneset is positively (red) or negatively (blue) skewed (w/ white as non-significant in the middle).
If this solution is not feasible, an alternative would be to do internally what is now shown in the vignette for fastGSEA, where the output is split into es>0 and es<0 and these results are reported separately. E.g., if the ranked signature sig is tested, its results should be split into two signatures, sig.up and sig.dn, and the hypeR-dot plot will have two corresponding "columns", so we'll know what's enriched up and what's enriched down.
Ideally, when hypeR is called with the KS test, the hypeR-dot plot should color-code the dots based on whether a geneset is positively (red) or negatively (blue) skewed (w/ white as non-significant in the middle).
If this solution is not feasible, an alternative would be to do internally what is now shown in the vignette for fastGSEA, where the output is split into es>0 and es<0 and these results are reported separately. E.g., if the ranked signature
sigis tested, its results should be split into two signatures,sig.upandsig.dn, and the hypeR-dot plot will have two corresponding "columns", so we'll know what's enriched up and what's enriched down.