I'm the beginner of bioinformatics.
In paper Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles, I don't know why the authors want to estimate the significance level of ES.
What is the null hypothesis here?
•
written
by
needonature •
With GSEA you're interested if some set of genes is, as a group, up/down regulated significantly more than one would expect by random chance. There's always an enrichment score (ES), but that doesn't mean we need to care about it.
Thanks!
But I'm not sure why we can use permutation test (I know this use in other situations well) here to get significance? Could you tell me more details?
•
written
by
needonature •
In the same method as every other permutation test, namely you permute group IDs and see how often you get as more or more enrichment. You'd be best off reading up on permutation tests on wikipedia or a statistics book.
Log in to answer this question.
