Genomic best linear unbiased prediction method including imprinting effects for genomic evaluation

Area of Science:

• Quantitative genetics
• Animal breeding
• Statistical genomics

Background:

• Genomic best linear unbiased prediction (GBLUP) assumes additive genetic effects.
• Genomic imprinting, a parent-of-origin-specific gene expression, contributes to genetic variation in quantitative traits.
• Existing GBLUP methods may not fully capture imprinting effects.

Purpose of the Study:

• To develop and evaluate two GBLUP models incorporating imprinting effects (GBLUP-I1 and GBLUP-I2).
• To assess the performance of these models in estimating variance components and predicting genetic values.
• To compare the accuracy of GBLUP-I models against standard GBLUP.

Main Methods:

• Developed GBLUP models based on genotypic values (GBLUP-I1) and gametic values (GBLUP-I2).
• Evaluated model performance using simulations across various genetic variation scenarios.
• Assessed estimation of variance components and prediction accuracy of total genetic values.

Main Results:

• GBLUP-I1 and GBLUP-I2 accurately estimated genetic and residual variances.
• Prediction accuracy increased with imprinting degree and heritability.
• GBLUP-I1 showed higher accuracy than GBLUP, especially when imprinting variance was 1.4%-6.0% of phenotypic variance.
• GBLUP-I2's superiority over GBLUP depended on imprinting degree and sex-specific genetic effects.

Conclusions:

• GBLUP-I1 and GBLUP-I2 accurately estimate total genetic variance and predict genetic values.
• GBLUP-I1 is recommended for general genomic evaluation.
• GBLUP-I2 is advantageous for traits with significant imprinting effects and sex-specific genetic differences.