Multi-trait multi-environment genomic prediction strategies for Miscanthus sacchariflorus

Area of Science:

• Plant Breeding
• Genomics
• Quantitative Genetics

Background:

• Genomic selection (GS) is crucial for enhancing genetic gain in complex traits within Miscanthus breeding programs.
• Developing improved Miscanthus cultivars necessitates evaluating genotypes across diverse environments for robust performance.
• Multi-trait multi-environment (MTME) genomic prediction (GP) models offer enhanced selection accuracy by integrating multiple data dimensions.

Purpose of the Study:

• To evaluate the predictive ability of five GP models, including three MTME models with genotype-by-trait-by-environment (G×E×T) interaction and two single-trait multi-environment (STME) models.
• To assess model performance across three realistic cross-validation schemes (predicting new genotypes, missing traits, and partially observed genotypes).

Main Methods:

• Analysis of a Miscanthus sacchariflorus population (336 genotypes) across three environments, evaluating four traits: biomass yield (YDY), total culm number (TCM), average internode length (AIL), and culm node number (CNN).
• Comparison of predictive abilities between MTME and STME models under various cross-validation scenarios.

Main Results:

• MTME models showed 10%–70% higher predictive ability than STME models for TCM and AIL across all cross-validation schemes.
• STME models performed similarly or better (5%–64%) than MTME models for YDY and CNN in most environments.
• MTME models improved prediction for genotypes untested across environments or lacking specific trait data, despite not outperforming STME for all traits.

Conclusions:

• MTME GP models can significantly enhance predictive ability for complex traits in Miscanthus, leading to shorter breeding cycles.
• Implementation of MTME models accelerates selection decisions and improves the development of superior Miscanthus cultivars.
• The choice between MTME and STME models depends on the specific trait and breeding objective, with MTME offering advantages for certain complex traits.