Skip to main content

Table 5 Inflation * of genomic predictions from different methods, obtained for five traits of a mice population

From: A comparison of statistical methods for genomic selection in a mice population

Splitting Method Trait
W6W WGS BL %CD8+ CD4+/CD8+
Within POL 1.21 0.91 0.89 1.03 0.97
emBayesB 1.43 1.57 3.70 1.65 1.55
RR_GBLUP 0.95 0.70 0.88 0.66 0.75
SS_BY 0.79 0.66 - 0.60 0.72
SS_ABS 0.72 0.49 0.49 0.64 0.72
RKHS 1.08 0.95 0.89 1.06 1.26
SVR 1.24 0.91 0.68 1.16 1.18
BayesCpi 0.55 0.29 0.40 0.17 0.29
BayesC 0.18 0.12 0.21 0.19 0.20
LASSO 0.93 1.01 1.08 1.00 1.07
RF 1.48 1.10 0.78 1.12 1.11
Across POL - - - - -
emBayesB 0.46 0.90 1.47 1.43 1.33
RR_GBLUP 0.37 0.49 0.52 0.46 0.52
SS_BY 0.30 0.46 - 0.42 0.52
SS_ABS 0.29 0.30 0.32 0.50 0.52
RKHS 0.55 0.95 0.75 1.21 1.36
SVR 0.61 0.89 0.62 1.42 1.27
BayesCpi 0.22 0.21 0.14 0.17 0.24
BayesC 0.07 0.08 0.11 0.18 0.15
LASSO 0.38 0.71 0.56 0.98 1.01
RF 0.72 1.22 0.65 1.27 1.22
  1. *Average of ten replicates. Inflation was measured as the slope of the regression of observed phenotypes on predicted phenotypes of testing set
  2. Trait = weight at 6 weeks (W6W), weight growth slope (WGS), body length (BL), percentage of CD8+ cells (%CD8+), ratio between CD4+ and CD8+ cells (CD4+/CD8+). Splitting= splitting strategy in cross-validation (within or across-family).