Table 6.

Performance comparison of deepGBLUP with the other genomic prediction methods on the simulated data across different heritabilities and single QTL effects

Heritability	Method	QTL effect
		a	d	e
0.5	GBLUP	0.633 ± 0.008	0.629 ± 0.008	0.613 ± 0.005
	DGBLUP	0.627 ± 0.008	0.624 ± 0.007	0.608 ± 0.005
	EGBLUP	0.630 ± 0.009	0.626 ± 0.008	0.611 ± 0.006
	BayesA	0.628 ± 0.01	0.622 ± 0.007	0.606 ± 0.006
	BayesB	0.626 ± 0.009	0.621 ± 0.008	0.602 ± 0.005
	BayesC	0.628 ± 0.009	0.625 ± 0.008	0.608 ± 0.005
	deepGBLUP	0.641 ± 0.007	0.635 ± 0.007	0.620 ± 0.006
0.3	GBLUP	0.588 ± 0.028	0.571 ± 0.026	0.566 ± 0.027
	DGBLUP	0.587 ± 0.029	0.571 ± 0.027	0.567 ± 0.027
	EGBLUP	0.587 ± 0.028	0.571 ± 0.026	0.565 ± 0.027
	BayesA	0.569 ± 0.027	0.552 ± 0.025	0.546 ± 0.026
	BayesB	0.583 ± 0.028	0.568 ± 0.026	0.564 ± 0.026
	BayesC	0.581 ± 0.028	0.567 ± 0.027	0.564 ± 0.027
	deepGBLUP	0.608 ± 0.028	0.594 ± 0.026	0.589 ± 0.026
0.1	GBLUP	0.457 ± 0.028	0.443 ± 0.023	0.433 ± 0.026
	DGBLUP	0.454 ± 0.028	0.441 ± 0.023	0.431 ± 0.026
	EGBLUP	0.462 ± 0.028	0.450 ± 0.023	0.439 ± 0.026
	BayesA	0.413 ± 0.031	0.388 ± 0.029	0.394 ± 0.033
	BayesB	0.446 ± 0.025	0.438 ± 0.025	0.415 ± 0.024
	BayesC	0.443 ± 0.029	0.439 ± 0.025	0.421 ± 0.029
	deepGBLUP	0.542 ± 0.023	0.532 ± 0.019	0.518 ± 0.022

Each value in the cells are means and standard errors of the predictive abilities for 10-fold tests. We highlight the best results in italic