Viral nervous necrosis resistance in gilthead sea bream (Sparus aurata) at the larval stage: heritability and accuracy of genomic prediction with different training and testing settings

Table 3 Accuracy of prediction of EBV for VNN symptomatology and β coefficient provided by the Bayesian models in three cross-validation strategies (random, K-means clustering and leave-one-family-out focused on the parents of the challenged fish)

Cross-validation strategy	Model	Accuracy	β
Random	Bayes B	0.903	1.010
	Bayes C	0.903	1.010
	Bayes Ridge Regression	0.903	1.010
K-means clustering	Bayes B	0.525	1.935
	Bayes C	0.530	1.956
	Bayes Ridge Regression	0.529	1.952
Leave-one-family-out	Bayes B	0.119	0.587
	Bayes C	0.126	0.620
	Bayes Ridge Regression	0.127	0.625

EBV breeding value estimated using the phenotypic information of the animal and of its full- and half-sibs
Accuracy Pearson product-moment correlation between EBV and genomic-predicted EBV
β slope of the regression of the observed on the predicted EBV

ISSN: 1297-9686