Functional enrichment / GSEA (pre-ranked)
Your data stays in your browser: files are read and processed on your device, never uploaded to any server, and cleared when you close the page. Free, no sign-in.
GSEA uses no significance cut-off on genes: it asks whether the members of a gene set cluster at the top (positive enrichment, NES > 0) or bottom (negative enrichment, NES < 0) of the ranked list. Use a signed metric such as the Wald statistic or t value and upload every measured gene.
Enrichment plot: the top panel is the running enrichment score, whose peak is the ES; the ticks mark where set members fall in the list; the bottom panel is the ranking metric. Leading-edge genes are the members before the peak (after it for negative enrichment) and usually drive the enrichment.
NES normalises ES for set size so sets can be compared; nominal P and FDR come from gene-set permutation (random sets of the same size) and vary slightly with the random seed. FDR < 0.25 is the usual exploratory threshold. A nominal P of 0 means less than 1/permutations.
This is computationally heavy: press “Run” after changing parameters.
Subramanian A. et al. (2005) PNAS 102:15545–15550; Mootha V. K. et al. (2003) Nat Genet 34:267–273. ES is the weighted Kolmogorov–Smirnov running sum (computed exactly); NES, nominal P and FDR use gene-set permutation as defined in GSEA. Example gene sets: Liberzon A. et al. (2015) Cell Systems 1:417–425.
Up to 60,000 genes; 1,000 permutations of 50 gene sets take a few seconds in the browser.
Send us your data and research question and you will receive a written plan within 1 working day: analysis steps, parameter rationale, deliverables and timeline. Quoted per project.