Genomic evaluation methods to include intermediate correlated features such as high-throughput or omics phenotypes.

Area of Science:

• Quantitative Genetics
• Genomic Prediction
• High-Throughput Phenotyping

Background:

• Gene expression and other high-throughput measurements (omics) serve as intermediates between DNA and phenotype.
• Omics data, despite measurement errors, are often heritable and can be more accessible than direct trait measurements.
• Developing methods to leverage omics data in genetic selection is crucial for improving breeding programs.

Purpose of the Study:

• To present methods and perspectives for genetic selection utilizing massively recorded intermediate omics traits.
• To introduce a hierarchical, integrated framework for prediction incorporating omics data.
• To develop a novel statistical model for genetic evaluation that accounts for omics-mediated and residual genetic effects.

Main Methods:

• Developed a two-nested genomic BLUP (Genomics Omics BLUP - GOBLUP) model.
• The first model estimates omics effects on traits and residual breeding values.
• The second model estimates omics-mediated breeding values, treating omics as heritable.

Main Results:

• GOBLUP allows for selection using omics measures, functioning as a single-step method.
• Heritability is decomposed into 'mediated' and 'not mediated' components, enabling prediction of omics accuracy.
• Even omics measures explaining small phenotypic variation can enhance selection accuracy if they are heritable.

Conclusions:

• GOBLUP provides a framework for integrating omics data into genetic evaluations, potentially increasing selection accuracy.
• The method allows for a nuanced understanding of genetic architecture by separating mediated and residual genetic effects.
• Further studies are needed to validate the practical utility of omics in selection on a case-by-case basis.