Elaheh Vojgani, Torsten Pook, Johannes W. R. Martini, Armin C. Hölker, Manfred Mayer, Chris-Carolin Schön & Henner Simianer

Theoretical and Applied Genetics September 2021; vol. 134: 2913–2930.

Key Message

The accuracy of genomic prediction of phenotypes can be increased by including the top-ranked pairwise SNP interactions into the prediction model.

Abstract

We compared the predictive ability of various prediction models for a maize dataset derived from 910 doubled haploid lines from two European landraces (Kemater Landmais Gelb and Petkuser Ferdinand Rot), which were tested at six locations in Germany and Spain. The compared models were Genomic Best Linear Unbiased Prediction (GBLUP) as an additive model, Epistatic Random Regression BLUP (ERRBLUP) accounting for all pairwise SNP interactions, and selective Epistatic Random Regression BLUP (sERRBLUP) accounting for a selected subset of pairwise SNP interactions. These models have been compared in both univariate and bivariate statistical settings for predictions within and across environments. Our results indicate that modeling all pairwise SNP interactions into the univariate/bivariate model (ERRBLUP) is not superior in predictive ability to the respective additive model (GBLUP). However, incorporating only a selected subset of interactions with the highest effect variances in univariate/bivariate sERRBLUP can increase predictive ability significantly compared to the univariate/bivariate GBLUP. Overall, bivariate models consistently outperform univariate models in predictive ability. Across all studied traits, locations and landraces, the increase in prediction accuracy from univariate GBLUP to univariate sERRBLUP ranged from 5.9 to 112.4 percent, with an average increase of 47 percent. For bivariate models, the change ranged from −0.3 to + 27.9 percent comparing the bivariate sERRBLUP to the bivariate GBLUP, with an average increase of 11 percent. This considerable increase in predictive ability achieved by sERRBLUP may be of interest for “sparse testing” approaches in which only a subset of the lines/hybrids of interest is observed at each location.

See: https://link.springer.com/article/10.1007/s00122-021-03868-1

Figure 1: Basic scheme of uni- and bivariate sERRBLUP across environments. All pairwise SNP interaction effects and their variances are estimated from all data in environment 1, and effects are ordered either by absolute effect size or effect variance (A). Then, an epistatic relationship matrix for all lines is constructed from the top ranked subset of interaction effects (B) which in the univariate model is used in environment 2 (C) to predict phenotypes of the test set (green) from the respective training set (red, D). In the bivariate model, this information is combined with the complete data from environment 1 (blue, E) to predict the test set.