Predicting inbred parent synchrony at flowering for maize hybrid seed production by integrating crop growth model with whole genome prediction

Abstract

One of the challenges of maize (Zea mays) hybrid seed production is to ensure synchrony at flowering of the two inbred parents of a hybrid, which depends on the specific parental combination and environmental conditions of the production field. Maize flowering can be simulated using a mechanistic crop growth model that converts thermal time accumulation to leaf numbers based on inbred-specific physiological parameter values. Heretofore, these inbred-specific physiological parameters need to be measured or assigned based on prior knowledge. Here, we leverage genetic, environmental, and management data to predict physiological parameters and simulate flowering phenotypes by using whole genome prediction methodology combined with a crop growth model (CGM–WGP) as part of in-field in-season inbred growth development. We use two estimation sets that differ in terms of management and weather information to test the robustness of our approach. As part of our findings, we demonstrate the importance of defining informative priors to generate biologically meaningful predictions of unobserved physiological parameters. Our CGM–WGP infrastructure is efficient at simulating flowering phenotypes. An important practical application of our method is the ability to recommend differential planting intervals for male and female maize inbreds used in commercial seed production fields to synchronize male and female flowering.