Accuracy of genotypic value predictions for marker-based selection in biparental plant populations

Abstract

The availability of cheap and abundant molecular markers has led to plant-breeding methods that rely on the prediction of genotypic value from marker data, but published information is lacking on the accuracy of genotypic value predictions with empirical data in plants. Our objectives were to (1) determine the accuracy of genotypic value predictions from multiple linear regression (MLR) and genomewide selection via best linear unbiased prediction (BLUP) in biparental plant populations; (2) assess the accuracy of predictions for different numbers of markers (N(M)) and progenies (N(P)) used in estimation; and (3) determine if an empirical Bayes approach for modeling of the variances of individual markers and of epistatic effects leads to more accurate predictions in empirical data. We divided each of four maize (Zea mays L.) datasets, one Arabidopsis dataset, and two barley (Hordeum vulgare L.) datasets into an estimation set, where marker effects were calculated, and a test set, where genotypic values were predicted based on markers. Predictions were more accurate with BLUP than with MLR. Predictions became more accurate as N(P) and N(M) increased, until sufficient genome coverage was reached. Modeling marker variances with the empirical Bayes method sometimes led to slightly better predictions, but the accuracy with different variants of the empirical Bayes method was often inconsistent. In nearly all cases, the accuracy with BLUP was not significantly different from the highest accuracy across all methods. Accounting for epistasis in the empirical Bayes procedure led to poorer predictions. We concluded that among the methods considered, the quick and simple BLUP approach is the method of choice for predicting genotypic value in biparental plant populations.

Prediction of genetic values of quantitative traits with epistatic effects in plant breeding populations

“…On one hand, Blanc et al (2006) suggested significant marker by background interactions in maize; Dudley (2008) and Dudley and Johnson (2009) argued for substantial benefits in using epistatic effects to improve genetic value prediction in intermated maize recombinant inbred lines; Hu et al (2011) gave an example for the advantages of using epistatic effects in soybean biparental populations. On the other hand, using nested mapping populations, the study in Buckler et al (2009) and Tian et al (2011) seemed to suggest very little role for the contribution of epistasis; Lorenzana and Bernardo (2009) found that including interaction terms in fact reduced prediction accuracy for the intermated recombinant inbred line population that they analyzed. More discussion on this topic can be found in Cooper et al (2009) and Lorenz et al (2011).…”

“…Significant literatures also exist in animal breeding research (for example, Gonzalez-Recio et al, 2008;van Raden et al, 2008;de los Campos et al, 2009a;Hayes et al, 2009 andToosi et al, 2009). In both settings, the large amount of variation resulted from hundreds or thousands of markers can be controlled by various shrinkage methods formulated in frequentist or Bayesian frameworks, which has notable success in multiple crop species (for example, Lorenzana andBernardo, 2009 andCrossa et al, 2010). In principle, epistatic effects can be incorporated just as main marker effects in these models, but the elevated number of epistatic effects can still pose serious problems when the number of markers is large.…”

Modeling Epistasis in Genomic Selection

“…Moreover, several genomic selection approaches have been developed to model both main and epistatic effects (Xu 2007;Cai et al 2011;Wittenburg et al 2011;Wang et al 2012). While in some studies prediction accuracies increased (Hu et al 2011), in others modeling epistasis adversely affected prediction accuracies (Lorenzana and Bernardo 2009).Despite these first attempts, epistasis is often ignored in genomic selection approaches using parametric models mainly because of the high associated computational load, especially if a large number of markers are available. An attractive solution to reduce the computational load is to extend genomic best linear unbiased prediction (G-BLUP) models (VanRaden 2008) by adding marker-based epistatic relationship matrices [extended genomic best linear unbiased prediction (EG-BLUP)].…”

“…Moreover, several genomic selection approaches have been developed to model both main and epistatic effects (Xu 2007;Cai et al 2011;Wittenburg et al 2011;Wang et al 2012). While in some studies prediction accuracies increased (Hu et al 2011), in others modeling epistasis adversely affected prediction accuracies (Lorenzana and Bernardo 2009).…”

Genomic Predictability of Interconnected Biparental Maize Populations

“…The first question has been investigated from a theoretical point of view (Daetwyler et al 2008(Daetwyler et al , 2010Goddard and Hayes 2009;Meuwissen 2009) and with low-density marker panels also empirically in maize (Lorenzana and Bernardo 2009). It is still unclear, however, how well theoretical expectations fit observed prediction accuracies in biparental maize populations, especially concerning their robustness across several traits and populations.…”