Genetic evaluation using single-step genomic best linear unbiased predictor in American Angus

Area of Science:

• Animal Genetics and Breeding
• Quantitative Genetics
• Genomic Prediction

Background:

• Traditional breeding values (EBV) rely on pedigree information and can be limited in accuracy.
• Genomic selection offers a powerful tool to enhance accuracy by utilizing single nucleotide polymorphism (SNP) data.
• Single-step genomic best linear unbiased prediction (ssGBLUP) integrates pedigree and genomic information for more precise EBV.

Purpose of the Study:

• To investigate the predictive ability of genomic EBV derived from ssGBLUP in Angus cattle.
• To compare ssGBLUP with traditional BLUP methods across various economically important traits.
• To evaluate the impact of reference population size and composition on prediction accuracy.

Main Methods:

• Utilized a large dataset with millions of records for birth weight (BiW), weaning weight (WW), postweaning gain (PWG), and calving ease (CE).
• Employed ssGBLUP and indirect prediction methods using SNP effects derived from different reference populations (ref_2k, ref_8k, ref_33k).
• Implemented cross-validation to assess predictive ability on a validation population and used an algorithm for proven and young animals (APY).

Main Results:

• ssGBLUP significantly improved predictivity for growth traits (BiW, WW, PWG) compared to BLUP, with gains increasing with larger reference populations (e.g., ref_33k).
• Predictivity for calving ease (CE) remained lower, attributed to its low incidence rate.
• Indirect predictions using SNP effects were as accurate as full ssGBLUP, and efficient algorithms (APY) provided substantial gains with reduced computational load.

Conclusions:

• Genomic evaluation using ssGBLUP is feasible in beef cattle, maintaining existing model complexities.
• Predictive gains are strongly influenced by the reference population's composition and size.
• Indirect genomic predictions enable accurate evaluations for young animals, especially with large reference populations, making ssGBLUP applicable to extensive genotyped populations.