Modeling additive and non-additive effects in a hybrid population using genome-wide genotyping: prediction accuracy...

Area of Science:

• Quantitative genetics
• Plant breeding
• Genomic selection

Background:

• Hybrids are vital in plant breeding, necessitating precise variance component estimation for optimizing genetic gain.
• Genome-wide information can enhance models for assessing additive and non-additive genetic variances and their prediction accuracy in genomic selection.

Purpose of the Study:

• To develop and compare linear mixed models using pedigree and marker-based relationships for estimating genetic effects (additive, dominance, epistatic).
• To evaluate the prediction accuracy of these models for genomic selection in Eucalyptus.
• To assess the performance of progeny models incorporating gametic phase information.

Main Methods:

• Ten linear mixed models were developed using parent-based pedigree and marker-based relationship matrices.
• Five progeny models utilizing gametic phase information were created to estimate additive (A), dominance (D), and epistatic (AA, AD, DD) effects.
• Models were compared using Eucalyptus tree height data, 3303 SNP markers, and metrics like AIC, variance ratios, goodness-of-fit, prediction accuracy, and MSE.

Main Results:

• Parent marker-based models outperformed pedigree-based models, showing better fit, accuracy, and lower AIC and MSE.
• Progeny gametic phase-based models demonstrated superior performance in fitting observations and predicting genetic values.
• Estimates for additive-by-dominance (AD) and dominance-by-dominance (DD) variances had high standard errors, and null estimates were found for AA and AD effects in progeny models.

Conclusions:

• Genome-wide information significantly improves the accuracy of variance component estimation and prediction in plant breeding.
• Progeny models incorporating gametic phase information are advantageous for predicting genetic values in hybrid breeding.
• Further research is needed to refine the estimation of complex epistatic effects.