Skip to main content
. 2010 Jan 18;11(Suppl 1):S18. doi: 10.1186/1471-2105-11-S1-S18

Figure 2.

Figure 2

The performance comparison between Lasso and AGL on pure epistasis model. Different main effects (additive, dominant and recessive) and their interactions are taken into consideration when fitting the Lasso model. The number in the x-axis is the number of SNPs simulated in the experiments. Lasso with a presumed model performs poorly with increasing number of noise SNPs, while AGL is robust under all settings.