Fig. 5
From: Empowering individual trait prediction using interactions for precision medicine
![Fig. 5](http://media.springernature.com/full/springer-static/image/art%3A10.1186%2Fs12859-021-04011-z/MediaObjects/12859_2021_4011_Fig5_HTML.png)
Performance in simulation scenario 8. Performance of the algorithms MBMDRC, RANGER, and GLMNET measured as AUC over 50 replicates in sample sizes 200, 1000, 2000, and 10,000 in scenario 8: three interacting SNPs without marginal effects and three SNPs with marginal effects only (MAF 0.1, 0.2, or 0.4 and heritability 0.05, 0.1, 0.2), 94 SNPs without any effect