| Summary: | In genomewide selection, the expected correlation between predicted performance and true genotypic value is a function of the training population size (), heritability on an entry-mean basis (), and effective number of chromosome segments underlying the trait (). Our objectives were to (i) determine how the prediction accuracy of different traits responds to changes in , , and number of markers () and (ii) determine if prediction accuracy is equal across traits if , , and are kept constant. In a simulated population and four empirical populations in maize ( L.), barley ( L.), and wheat ( L.), we added random nongenetic effects to the phenotypic data to reduce to 0.50, 0.30 and 0.20. As expected, increasing , , and increased prediction accuracy. For the same trait within the same population, prediction accuracy was constant for different combinations of and that led to the same . Different traits, however, varied in their prediction accuracy even when , , and were constant. Yield traits had lower prediction accuracy than other traits despite the constant , , and . Empirical evidence and experience on the predictability of different traits are needed in designing training populations.
|
|---|
