TABLE 2.

Prediction accuracy of SVR, KRR, ENET, GBLUP, and BayesB for the three datasets.

Dataset	Trait	SVR	KRR	ENET	GBLUP	BayesB
Beef cattle	LW	0.274 ± 0.022	0.283 ± 0.019	0.276 ± 0.018	0.256 ± 0.017	0.265 ± 0.016
	CW	0.307 ± 0.016	0.315 ± 0.015	0.315 ± 0.017	0.292 ± 0.014	0.282 ± 0.012
	EMA	0.280 ± 0.025	0.281 ± 0.022	0.285 ± 0.024	0.292 ± 0.015	0.281 ± 0.015
Dairy cattle	MY	0.764 ± 0.013	0.781 ± 0.009	0.762 ± 0.014	0.768 ± 0.006	0.767 ± 0.005
	MFP	0.796 ± 0.012	0.828 ± 0.006	0.797 ± 0.012	0.832 ± 0.003	0.855 ± 0.003
	SCS	0.706 ± 0.010	0.751 ± 0.008	0.722 ± 0.019	0.752 ± 0.006	0.731 ± 0.003
Loblolly pine	HT	0.340 ± 0.027	0.352 ± 0.011	0.366 ± 0.014	0.349 ± 0.012	0.365 ± 0.009
	CWAL	0.352 ± 0.022	0.359 ± 0.018	0.369 ± 0.022	0.384 ± 0.014	0.400 ± 0.011
	TS	0.397 ± 0.017	0.407 ± 0.016	0.398 ± 0.015	0.366 ± 0.012	0.418 ± 0.013

The accuracy was calculated by the Pearson’s correlation. LW, live weight; CW, carcass weight; EMA, eye muscle area; MY, milk yield; MFP, milk fat percentage; SCS, somatic cell score; HT, total stem height; CWAL, crown width along the planting beds; TS, tree stiffness. SVR, support vector regression; KRR, kernel ridge regression; ENET, elastic net; GBLUP, genomic best linear unbiased prediction. The bold values mean the highest prediction accuracy for each trait.