Including dominance effects in the genomic BLUP method for genomic evaluation

Area of Science:

• Animal Genetics
• Quantitative Genetics
• Genomic Selection

Background:

• Genomic Best Linear Unbiased Prediction (GBLUP) is a key tool for genetic merit estimation.
• Incorporating dominance effects can potentially enhance prediction accuracy in crossbred populations.
• The impact of dominance variance on GBLUP performance requires further investigation.

Purpose of the Study:

• To evaluate the performance of GBLUP including dominance genetic effects (GBLUP-D).
• To assess GBLUP-D's accuracy in estimating genetic variances and predicting genetic merits.
• To compare GBLUP-D with traditional GBLUP using simulation and real pig data.

Main Methods:

• Computer simulations were used to model genetic effects.
• GBLUP-D was applied to estimate variances and predict genetic merits.
• Performance was evaluated using simulation data and two actual pig traits (T4 and T5).

Main Results:

• GBLUP-D explained over 50% of dominance genetic variance in simulations.
• GBLUP-D improved the accuracy of estimated total genetic effects by over 1.2% compared to GBLUP.
• Accuracy gains were more substantial with larger dominance variances and increased marker numbers.
• Small dominance variances in pig traits T4 (9.6%) and T5 (6.3%) showed minimal improvement in prediction accuracy.
• Additive genetic effect estimates were similar between GBLUP-D and GBLUP.

Conclusions:

• GBLUP-D is a feasible method for improving genetic performance in crossbred populations with substantial dominance variation.
• GBLUP-D can aid in identifying mating systems with superior combining ability.
• The effectiveness of GBLUP-D is contingent on the magnitude of dominance genetic variance.