Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: Roastgsa: a comparison of rotation-based scores for gene set enrichment analysis

Fig. 2

Characteristics of all presented scores: performance in simulated data is measured from 1 (poor) to 10 (great) based on the obtained recovery rates (the average recovery rate relative to the best rate); performance in benchmarking data is measured from 1 to 10 based on the M1 ranking; computational time is measured relative to the fastest method; Scores that were implemented in limma are specified for both romer (competitive scores) and roast (self-contained scores) functions

Back to article page