Updated knowledge in the estimation of genetics parameters: a Bayesian approach in white oat (Avena sativa L.)

The present study aimed to implement a Bayesian framework for genetic analysis in crop species breeding and present different procedures for informative prior elicitation. Genetic Bayesian estimation in crop breeding was performed using ten years of evaluation in the white oat population. The Bayesian framework was based on the MCMC Generalized Linear Mixed Models (MCMCglmm) R-package. The procedure for updating knowledge was performed. For the first year, a non-informative prior for the variance components was used. For the second year, a non-informative prior and an informative prior defined as the posterior from the non-informative and informative analyses of the first year were used. And in the following years, prior distributions were similarly built. The inference results were compared using the posterior coefficient of variation (CV) of the estimates of the components of variance, heritability and additive genetic values and length of the Highest Posterior Density intervals (HPD) of the parameter estimates. The informative prior presented smaller values of CV of the posterior densities, and the length of the HPD interval was also smaller. Thus, the updating procedures will provide breeders with the knowledge accumulated over the years.