Putting genes into genetic coefficients

Plant parameters are critical inputs in crop simulation models and allow a general set of algorithms to represent features of specific cultivars. A subset of plant parameters is often referred to as "genetic coefficients". However, these genetic coefficients are developed from phenotypic observations, usually have a weak genetic basis, and are at best "genotypic" coefficients because they consider the genotype from a very integrative perspective and likely include some impact of environment on the trait or characteristic described. With increased understanding of crop genomes, we believe models can be improved by incorporating genetic coefficients that accurately describe the action of specific genes (within the genome) and therefore better represent the association between gene function and plant phenotype in simulation models. As an example, we discuss how knowledge of height genes in wheat (Triticum aestivum L.) cultivars, along with stronger genetic and environmental response algorithms, could substitute for the phenotypic parameter "height class" in the model SHOOTGRO. We also demonstrate how models containing responses based on known genetic variation can be used to identify traits to incorporate into cultivars better adapted to future climate scenarios. It remains for the geneticist, plant breeder, physiologist and modeler to cooperate and communicate with each other so that genetic information and responses with the genotype and environment and their interaction can be described in models and used to develop cultivars better able to exploit future climatic conditions. (C) 2004 Elsevier B.V. All rights reserved.