Skip to main content

Table 5 Average estimated variance components (standard deviation in brackets) and average accuracy ρ of genetic value prediction*

From: Including non-additive genetic effects in Bayesian methods for the prediction of genetic values based on genome-wide markers

Simulation without epistasis

Method

Model

σ a 2

σ d 2

σ a a 2

σ a d 2

σ d a 2

σ d d 2

σ e 2

ρ

BayesB

M1

0.631

0.056

-

-

-

-

0.652

0.860

  

(0.204)

(0.035)

    

(0.180)

 

fBayesB

M1

0.699

0.165

-

-

-

-

0.413

0.760

  

(0.207)

(0.065)

    

(0.132)

 

fBayesB

M2

0.732

0.304

0.036

0.065

0.068

0.074

0.170

0.608

  

(0.214)

(0.112)

(0.028)

(0.036)

(0.042)

(0.046)

(0.066)

 

Simulated components

0.709

0.043

-

-

-

-

0.754

-

Simulation with epistasis

Method

Model

σ a 2

σ d 2

σ a a 2

σ a d 2

σ d a 2

σ d d 2

σ e 2

ρ

BayesB

M1

0.949

0.215

-

-

-

-

1.968

0.585

  

(0.250)

(0.067)

    

(0.266)

 

fBayesB

M1

0.920

0.267

-

-

-

-

1.567

0.543

  

(0.197)

(0.080)

    

(0.282)

 

fBayesB

M2

1.277

0.910

0.171

0.275

0.296

0.305

0.493

0.340

  

(0.230)

(0.269)

(0.086)

(0.106)

(0.127)

(0.126)

(0.257)

 

Simulated components

1.284

0.192

0.308

0.103

0.071

0.014

1.952

-

  1. * 230-QTL scenario with 5 227 markers and H2 = 0.5. Variance components for each source of genetic variation: σ a 2 additive genetic, σ d 2 dominance, σ a a 2 additive × additive, σ a d 2 additive × dominance, σ d a 2 dominance × additive, σ d d 2 dominance × dominance, residual variance σ e 2 . M1 includes additive and dominance effects, M2 includes additive, dominance and pairwise epistatic effects.