how to calculate running sum in R for GSEA

Question

[Mo]

Hello,

I am trying to implement a R function which does the GSEA.

I read many papers related to this method and each of them tries to destroy the other and show a better performance of its own method (that is what we do as scientists :-D )

Anyway, what I am now working on is to find out how running sum works to calculate the Score!

The running sum is to calculate the Enrichment Score over a gene set

1. How to define a gene set? For example if I have over 20000 genes, can I say the first 200 are one set, and the rest is another set?
2. How to calculate it ? what they say

a Kolmogorov-Smirnov (K-S) running sum statistic is computed: beginning with the top-ranking gene, the running sum increases when a gene annotated to be a member of gene set S is encountered and decreases otherwise

Can someone explain how does this technique work?

Can it be done for one sample? If not, why?

[maciek.kandula]

You can actually download and have a look at how the authors implemented the original GSEA in R: http://www.broadinstitute.org/cancer/software/gsea/wiki/index.php/R-GSEA_Readme

Answer 1

[Giovanni M Dall'Olio]

The GSEA algorithm is already implemented in other R libraries. For example you can check the functions gseGO or gseAnalyzer from the clusterProfiler package. Check also the vignette for more examples.