Bayesian optimization of multivariate genomic prediction models based on secondary traits for improved accuracy gains and phenotyping costs

In recent years, the genomic prediction that predicts phenotypic values from marker genotype data has attracted much more attention in the area of breeding. Especially, genomic selection using prediction values based on genomic prediction models has been contributing to more efficient and rapid breeding. In genomic prediction, it is important to construct the prediction model so that its accuracy becomes higher. Thus, multivariate genomic prediction models with secondary traits, such as data from various omics technologies including high-throughput phenotyping (e.g., unmanned aerial vehicle-based remote sensing), have started to be applied to many datasets because it offers improved accuracy gains compared with genomic prediction based only on marker genotypes. Although there is a trade-off between accuracy gains and phenotyping costs of secondary traits, no attempt has been made to optimize these trade-offs. In this study, we propose a novel approach to optimize multivariate genomic prediction models with secondary traits measurable at early growth stages for improved accuracy gains and phenotyping costs. The proposed approach employs Bayesian optimization for efficient Pareto frontier estimation, representing the maximum accuracy at a given cost. The proposed approach successfully estimated the optimal secondary trait combinations across a range of costs while providing genomic predictions for only about 20% of all possible combinations. The simulation results reflecting the characteristics of each scenario of the simulated target traits showed that the obtained optimal combinations were reasonable. Analysis of real-time target trait data showed that the proposed multivariate genomic prediction model had significantly superior accuracy compared to the univariate genomic prediction model.