Table 4.

Prediction performance in soybean for five traits of eight methods in terms of Pearson’s correlation (taken from Liu et al., 2019); A). Methods dualCNN, deepGS and singleCNN are different versions of CNN. Prediction performance in terms of Average Spearman Correlation (ASC) and mean square error (MSE) with genotype × environment interaction (I) and without genotype × environment interaction (WI) in a wheat dataset for trait Fusarium head blight (FHB) severity data (Montesinos-López et al., 2020; B)

A	Method	Yield	Protein	Oil	Moisture	Height
	dualCNN	0.452	0.619	0.668	0.463	0.615
	DeepGS	0.391	0.506	0.531	0.31	0.452
	Dense	0.449	0.603	0.657	0.427	0.612
	singleCNN	0.463	0.573	0.627	0.449	0.565
	rrBLUP	0.412	0.392	0.39	0.413	0.458
	BRR	0.422	0.392	0.39	0.413	0.458
	Bayes A	0.419	0.393	0.388	0.415	0.458
	BL	0.419	0.394	0.388	0.416	0.458
B	Interaction	Type	ASC	SE	MSE	SE
	I	BRR	0.584	0.012	3.015	0.169
	I	NDNN	0.626	0.013	1.891	0.088
	I	GP	0.596	0.01	2.457	0.121
	I	PDNN	0.627	0.012	1.912	0.073
	WI	BRR	0.436	0.018	4.481	0.25
	WI	NDNN	0.635	0.013	1.872	0.084
	WI	GP	0.431	0.018	3.418	0.186
	WI	PDNN	0.584	0.014	2.853	0.412