Quantitative prediction of mouse class I MHC peptide binding affinity using support vector machine regression (SVR) models

Table 2 Comparison between the additive method and the SVRMHC method in models constructed with the H2-Db, H2-Kb and H2-Kk datasets.

	H2-Db		H2-Kb		H2-Kk
	Additive Method	SVRMHC Method	Additive Method	SVRMHC Method	Additive Method	SVRMHC Method
Numbers of outliers	6	3	7	6	2	0
r ² (self-consistency, outliers removed)	0.946	0.983	0.989	0.97	0.933	0.973
r ² (self-consistency, entire dataset)	0.602	0.749	0.37	0.568	0.849	0.973
Average Residual (entire dataset)	0.403	0.17	0.443	0.382	0.178	0.039
Average Residual (outliers removed)	0.187	0.043	0.095	0.13	0.151	0.039
q ² (LOO_CV) (outliers removed)	0.401	0.456	0.454	0.486	0.456	0.721

ISSN: 1471-2105