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

BMC Bioinformatics

Table 4 Accuracy and bias of predicted GBVs evaluated with cross-validation in Data I

Method			Trait A	Trait B	Trait C
MCBayes	π = 0	r _TBV,pGBV	0.832 ± 0.039	0.611 ± 0.095	0.501 ± 0.083
		b _TBV,pGBV	1.016 ± 0.022	1.072 ± 0.231	1.052 ± 0.341
		r _y,pGBV	0.741 ± 0.037	0.191 ± 0.045	0.160 ± 0.050
		b _y,pGBV	1.013 ± 0.015	1.062 ± 0.125	1.039 ± 0.130
	0 < π <1	r _TBV,pGBV	0.791 ± 0.048	0.603 ± 0.112	0.390 ± 0.127
		b _TBV,pGBV	1.132 ± 0.064	1.210 ± 0.303	1.180 ± 0.379
		r _y,pGBV	0.705 ± 0.045	0.191 ± 0.046	0.121 ± 0.060
		b _y,pGBV	1.131 ± 0.065	1.210 ± 0.170	1.119 ± 0.373
varBayes	π = 0	r _TBV,pGBV	0.813 ± 0.049	0.620 ± 0.108	0.470 ± 0.118
		b _TBV,pGBV	1.080 ± 0.048	0.994 ± 0.157	0.963 ± 0.201
		r _y,pGBV	0.722 ± 0.047	0.187 ± 0.056	0.143 ± 0.058
		b _y,pGBV	1.072 ± 0.049	0.945 ± 0.195	0.931 ± 0.289
	0 < π <1	r _TBV,pGBV	0.779 ± 0.059	0.593 ± 0.111	0.423 ± 0.123
		b _TBV,pGBV	0.944 ± 0.040	0.816 ± 0.139	0.662 ± 0.153
		r _y,pGBV	0.690 ± 0.056	0.180 ± 0.055	0.125 ± 0.062
		b _y,pGBV	0.935 ± 0.039	0.787 ± 0.166	0.626 ± 0.255
single-trait	π = 0	r _TBV,pGBV	0.826 ± 0.040	0.515 ± 0.073	0.505 ± 0.074
(MCBayes)		b _TBV,pGBV	0.997 ± 0.023	1.073 ± 0.270	1.030 ± 0.303
		r _y,pGBV	0.735 ± 0.039	0.159 ± 0.045	0.162 ± 0.044
		b _y,pGBV	0.993 ± 0.012	1.071 ± 0.162	1.055 ± 0.222
	0 < π <1	r _TBV,pGBV	0.821 ± 0.039	0.531 ± 0.099	0.522 ± 0.094
		b _TBV,pGBV	1.131 ± 0.046	1.265 ± 0.555	1.192 ± 0.405
		r _y,pGBV	0.731 ± 0.037	0.164 ± 0.051	0.164 ± 0.048
		b _y,pGBV	1.127 ± 0.043	1.249 ± 0.482	1.205 ± 0.457

Averages and standard errors evaluated with 10- fold cross-validation are listed based on 100 replicates of simulated data in Data I are listed for prediction accuracy, r_pGBV,TBV, and bias, b_pGBV,TBV, as well as correlation between phenotypic value and predicted GBV, r_y,pGBV, and regression of phenotypic value on predicted GBV, b_y,pGBV , of each trait. For the settings of prior probability that a SNP has zero effect, π, see Table1.

ISSN: 1471-2105