Imitating the 'breeder's eye': predicting grain yield from measurements of non-yield traits

Abstract

Plant breeding relies on information gathered from field trials to select promising new crop varieties for release to farmers and to develop genomic prediction models that can enhance the efficiency and rate of genetic improvement in future breeding cycles. However, field trials conducted in one environment provide limited insight into how well crop varieties will perform in other environments. As the pace of climate change intensifies, the time lag of developing and deploying new crop varieties indicates that plant breeders will need to make decisions about new crop varieties without knowing the future environments those crop varieties will encounter in farmers fields. Therefore, significant improvements in cross-environment prediction of crop performance are essential for creating and maintaining resilient agricultural systems in the latter half of the twenty-first century. To address this challenge, we conducted linked yield trials of 752 public maize genotypes in two distinct environments: Lincoln, Nebraska, and East Lansing, Michigan. Our findings confirmed that genomic predictions of yield can outperform direct yield measurements used to train the genomic prediction model in predicting yield in a second environment. Additionally, we developed and trained another trait-based yield prediction model, which we refer to as the Silicon Breeders Eye (SBE). Our results demonstrate that SBE prediction has comparable predictive power to genomic prediction models. SBE prediction has the potential to be applied to a wider range of breeding programs, including those that lack the resources to genotype large populations of individuals, such as programs in the developing world, breeding programs for specialty crops, and public sector programs.