A Bayesian method and its variational approximation for prediction of genomic breeding values in multiple traits

BMC Bioinformatics

Table 1 Accuracy and bias of predicted GBVs in Data I

Method			Trait A	Trait B	Trait C
MCBayes	π = 0	r _TBV,pGBV	0.788 ± 0.051	0.581 ± 0.103	0.453 ± 0.090
		b _TBV,pGBV	0.994 ± 0.038	1.048 ± 0.264	1.00 ± 0.370
	0 < π <1	r _TBV,pGBV	0.753 ± 0.060	0.580 ± 0.117	0.364 ± 0.137
		b _TBV,pGBV	1.070 ± 0.064	1.149 ± 0.340	1.016 ± 0.364
varBayes	π = 0	r _TBV,pGBV	0.754 ± 0.061	0.570 ± 0.113	0.383 ± 0.117
		b _TBV,pGBV	1.054 ± 0.051	0.994 ± 0.233	0.899 ± 0.247
	0 < π <1	r _TBV,pGBV	0.716 ± 0.070	0.548 ± 0.122	0.347 ± 0.131
		b _TBV,pGBV	0.894 ± 0.054	0.834 ± 0.186	0.636 ± 0.202
single-trait	π =0	r _TBV,pGBV	0.783 ± 0.051	0.469 ± 0.083	0.455 ± 0.076
(MCBayes)		b _TBV,pGBV	0.978 ± 0.037	1.020 ± 0.301	0.970 ± 0.259
	0 < π <1	r _TBV,pGBV	0.778 ± 0.050	0.491 ± 0.114	0.483 ± 0.101
		b _TBV,pGBV	1.089 ± 0.054	1.110 ± 0.634	1.061 ± 0.338

Averages and standard errors based on 100 replicates of simulated data are listed for prediction accuracy, r_pGBV,TBV, and bias, b_pGBV,TBV, of each trait. For the prior probability that a SNP has zero effect, π, we considered two settings, in which π was fixed at 0, meaning the inclusion of all SNPs in the model, and π was varied over 0 < π <1 and inferred from the data.

ISSN: 1471-2105