Why Accuracy Metrics Fall Short in Comparing Phenomic and Genomic Prediction Models

Phenomic Selection is a new paradigm in plant breeding that uses high-throughput phenotyping technologies and machine learning models to predict traits of new individuals and make selections. This can allow breeders to evaluate more plants in higher throughput more accurately, resulting in faster rates of gain and reduced labor costs. However, Phenomic Prediction models are frequently benchmarked against Genomic Prediction models using cross-validation to demonstrate their usefulness to breeders. We argue that this is inappropriate for two reasons: 1) Differences in the accuracy statistic measured by cross-validation do not reliably indicate differences in the accuracy parameter of the breeder’s equation, and 2) Accuracy alone is insufficient to compare breeding schemes using Phenomic vs. Genomic Prediction because these tools differentially influence other parameters of the breeder’s equation. We show analytically and through re-analysis of data from three representative Phenomic Prediction studies that conclusions about the superiority of Phenomic Prediction over Genomic Prediction change if compared using consistent methods. We conclude that Phenotypic Selection may be useful, but comparisons of accuracy between Genomic Prediction and Phenotypic Prediction models are not.