Table 3. Predictive abilities obtained with LASSO, the elastic net, BayesB, and RR-BLUP for three experimental data sets.

			Predictive ability
Data	Trait	n	LASSO	Elastic net	BayesB	RR-BLUP
Rice	Flowering time	359	0.46 ± 0.013	0.52 ± 0.011	0.59 ± 0.006	0.59 ± 0.007
	Plant height	383	0.71 ± 0.005	0.73 ± 0.003	0.76 ± 0.002	0.76 ± 0.002
	Panicle length	375	0.60 ± 0.004	0.61 ± 0.007	0.66 ± 0.006	0.66 ± 0.006
	Seed length	377	0.75 ± 0.003	0.78 ± 0.003	0.75 ± 0.003	0.75 ± 0.003
Wheat	Yield	254	0.44 ± 0.010	0.43 ± 0.006	0.51 ± 0.006	0.51 ± 0.005
	TKWa	254	0.48 ± 0.010	0.48 ± 0.011	0.55 ± 0.004	0.55 ± 0.004
	Days to heading	254	0.60 ± 0.005	0.61 ± 0.008	0.66 ± 0.005	0.66 ± 0.004
Arabidopsis	Flowering time	180	0.71 ± 0.008	0.65 ± 0.004	0.75 ± 0.004	0.75 ± 0.004
	Plant diameter	180	0.47 ± 0.012	0.49 ± 0.016	0.51 ± 0.006	0.51 ± 0.007
	FRIb	164	0.48 ± 0.020	0.36 ± 0.026	0.41 ± 0.014	0.41 ± 0.013
	Plant width	176	0.46 ± 0.024	0.49 ± 0.010	0.44 ± 0.009	0.43 ± 0.009

Predictive abilities for all 11 traits were estimated using fivefold cross-validation and results were averaged over 10 replications ± SE.

^aTKW, thousand-kernel weight.

^bFRI, FRI gene expression.