|Wu, Jixiang - MISSISSIPPI STATE UNIV|
|Wu, Dongfeng - MISSISSIPPI STATE UNIV|
Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: July 12, 2005
Publication Date: January 1, 2006
Citation: Wu, J., Jenkins, J.N., McCarty Jr., J.C., Wu, D. 2006. Variance component estimation using the additive, dominance, and additive x additive model when genotypes vary across environments. Crop Science. 46:174-179. Interpretive Summary: When evaluating genetic and plant breeding experiments it is often necessary to plant seed from two or more generations plus parents in multiple environments. Practical difficulties often prevent the production of enough seed from each generation to plant in all environments. Methodologies are needed to utilize partial designs where data from all generations are not available from each environment. We have used an ADAA genetic model and simulated six genetic designs each with various combinations of generations in two environments. We conducted 500 Monte Carlo simulations with each design and report simulation results which allows a scientist to determine the relative bias of additive, dominance, and interaction genetic effects among the partial genetic designs. Genetic variance components for additive effects, additive by environment, and A x A effects were estimated and were unbiased for all six designs. Variance components for dominance effects and dominance by environment effects could be estimated but with large bias for two of the six designs and were largely unbiased for the other four designs. These results should be useful to geneticists and plant breeders conducting research across environments with more than one generation of seed. These results allow the scientist to choose how to best allocate experimental units among environments to estimate genetic effects with the least amount of bias.
Technical Abstract: In addition to additive and dominant genetic effects, the additive by additive interaction (or AA epistatic) effects that control many quantitative traits are important for genetic and breeding studies. To estimate these genetic variance components including GxE interaction one usually expects to have data from at least two generations (i.e., F1 and F2) and parents with the same entries in all environments. Practical difficulties may arise in implementing such a design. In this study, we performed Monte Carlo simulations to compare the estimated variance components between four partial and two complete genetic designs using the mixed lienar model approach. Our definition for genetic design is different from the traditional definitions of genetic mating designs. Simulation results showed that the estimated genetic variance components for additive, additive x environment, AA epistatic, and AA x environment effects were unbiased for the six designs. Among four partial designs, two provided the comparable results, for dominance and dominance x environment effects, to the complete genetic designs but with slightly larger mean square errors, indicating that some partial genetic designs can be used when the genetic resources are limited.