Skip to main content

Table 10 The prediction performance evaluated by the Area under curve (AUC) of GBLUP, BayesR and HyB_BR on seven diseases

From: A hybrid expectation maximisation and MCMC sampling algorithm to implement Bayesian mixture model based genomic prediction and QTL mapping

Diseases GBLUP BayesR HyB_BR
AUC h 2 AUC h 2 AUC h 2
BD 0.63(0.0135) 0.71 0.63(0.0131) 0.63 0.64(0.0174) 0.63
CAD 0.58(0.0116) 0.38 0.59(0.0118) 0.38 0.58(0.0131) 0.38
CD 0.60(0.0134) 0.69 0.65(0.0159) 0.61 0.65(0.0158) 0.61
HT 0.58(0.0125) 0.53 0.58(0.0131) 0.52 0.58(0.0140) 0.51
RA 0.58(0.0109) 0.50 0.70(0.0104) 0.45 0.70(0.0107) 0.45
T1D 0.64(0.0133) 0.66 0.86(0.0099) 0.63 0.86(0.0102) 0.63
T2D 0.59(0.0139) 0.59 0.60(0.0117) 0.52 0.60(0.0122) 0.52
  1. the heritability (h 2) is estimated by the equation \( {h}^2=\raisebox{1ex}{${\upsigma}_{\mathrm{g}}^2$}\!\left/ \!\raisebox{-1ex}{$\left({\upsigma}_{\mathrm{e}}^2+{\upsigma}_{\mathrm{g}}^2\right)$}\right. \); σ 2e is derived separately by three methods; fixed genetic variance of σ 2g for BayesR and HyB_BR is obtained from GCTA