A0305
Title: Regression approaches for modelling genotype-environment interaction and making predictions into unseen environments
Authors: Maksym Hrachov - University of Hohenheim (Germany) [presenting]
Hans-Peter Piepho - Universitaet Hohenheim (Germany)
Niaz Md Farhat Rahman - Bangladesh Rice Research Institute (Bangladesh)
Waqas Ahmed Malik - University of Hohenheim (Germany)
Abstract: In plant breeding and variety testing, there is an increasing interest in making use of environmental information to enhance predictions for new environments. Linear mixed models that have been proposed for this purpose are reviewed, with an emphasis on predictions and on methods to assess the uncertainty of predictions for new environments. The point of departure is straight-line regression, which may be extended to multiple environmental covariates and genotype-specific responses. When observable environmental covariates are used, this is also known as factorial regression. Early work along these lines can be traced back to Finlay-Wilkinson regression dating back to 1930s. This method, in turn, has close ties with regression on latent environmental covariates and factor-analytic variance-covariance structures for genotype-environment interaction. Extensions of these approaches - reduced rank regression, kernel- or kinship-based approaches, random coefficient regression, and extended Finlay-Wilkinson regression - are the focus of comparison. The objective is to demonstrate how seemingly disparate methods are very closely linked and fall within a common model-based prediction framework. Options are considered for assessing uncertainty of predictions, including cross-validation and model-based estimates of uncertainty, and tested methods on the long-term rice variety trial dataset.
