Managing Repeated Measures in the Analysis of Breeding Trials

We had an insightful webinar with Dr. Salvador Gezan, where he discussed the analysis of complex datasets derived from breeding trials. The webinar focused on the challenges posed by repeated measures, where multiple observations were taken for each plot throughout the trial.
During the webinar, we explored the use of ASReml-R (vsni.co.uk/software/asreml-r) and learned the essential steps for thorough data checking and the effective definition of a linear mixed model capable of handling correlated observations. The correlation present in both the errors and genotypes was addressed, ensuring a comprehensive understanding of the model's construction.
Furthermore, we delved into different variance-covariance structures, such as autoregressive, exponential, and uniform correlation. We examined their properties and practical applications, providing practical insights to help attendees choose the appropriate structure for their own analyses. This empowered participants to make informed decisions in their breeding programs.
The webinar also covered obtaining valuable summary statistics for each genotype within the model. Techniques like predictions, BLUEs, and BLUPs were explained, offering assistance in selecting the best genotypes for breeding programs.
The webinar was designed for individuals interested in analyzing breeding trial data, including researchers, plant breeders, and anyone curious about the subject. It provided a wealth of information and practical knowledge in this field.