Simon Rio, Deniz Akdemir, Tiago Carvalho & Julio Isidro y Sánchez

Theoretical and Applied Genetics; February 202; vol. 135: 405–419

Key message

New forms of the coefficient of determination can help to forecast the accuracy of genomic prediction and optimize experimental designs in multi-environment trials with genotype-by-environment interactions.

Abstract

In multi-environment trials, the relative performance of genotypes may vary depending on the environmental conditions, and this phenomenon is commonly referred to as genotype-by-environment interaction (G××E). With genomic prediction, G××E can be accounted for by modeling the genetic covariance between trials, even when the overall experimental design is highly unbalanced between trials, thanks to the genomic relationship between genotypes. In this study, we propose new forms of the coefficient of determination (CD, i.e., the expected model-based square correlation between a genetic value and its corresponding prediction) that can be used to forecast the genomic prediction reliability of genotypes, both for their trial-specific performance and their mean performance. As the expected prediction reliability based on these new CD criteria is generally a good approximation of the observed reliability, we demonstrate that they can be used to optimize multi-environment trials in the presence of G××E. In addition, this reliability may be highly variable between genotypes, especially in unbalanced designs with complex pedigree relationships between genotypes. Therefore, it can be useful for breeders to assess it before selecting genotypes based on their predicted genetic values. Using a wheat population evaluated both for simulated and phenology traits, and two maize populations evaluated for grain yield, we illustrate this approach and confirm the value of our new CD criteria.

See https://link.springer.com/article/10.1007/s00122-021-03972-2

Figure 1: Diagram illustrating the procedure to allocate candidate set genotypes to MET experimental designs (Opt, I, C1, and C2) and predict the genetic value of test set individuals. Genotypes are allocated at random for designs I, C1, and C2, while they are allocated using an optimization approach based on CD criteria for the Opt design. Note that optimized designs do not allow for a genotype to be replicated within each trial, but they do allow a genotype to be replicated across trials