Beyond Genomic Prediction: Combining Different Types of omics Data Can Improve Prediction of Hybrid Performance in Maize

Abstract

The ability to predict the agronomic performance of single-crosses with high precision is essential for selecting superior candidates for hybrid breeding. With recent technological advances, thousands of new parent lines, and, consequently, millions of new hybrid combinations are possible in each breeding cycle, yet only a few hundred can be produced and phenotyped in multi-environment yield trials. Well established prediction approaches such as best linear unbiased prediction (BLUP) using pedigree data and whole-genome prediction using genomic data are limited in capturing epistasis and interactions occurring within and among downstream biological strata such as transcriptome and metabolome. Because mRNA and small RNA (sRNA) sequences are involved in transcriptional, translational and post-translational processes, we expect them to provide information influencing several biological strata. However, using sRNA data of parent lines to predict hybrid performance has not yet been addressed. Here, we gathered genomic, transcriptomic (mRNA and sRNA) and metabolomic data of parent lines to evaluate the ability of the data to predict the performance of untested hybrids for important agronomic traits in grain maize. We found a considerable interaction for predictive ability between predictor and trait, with mRNA data being a superior predictor for grain yield and genomic data for grain dry matter content, while sRNA performed relatively poorly for both traits. Combining mRNA and genomic data as predictors resulted in high predictive abilities across both traits and combining other predictors improved prediction over that of the individual predictors alone. We conclude that downstream "omics" can complement genomics for hybrid prediction, and, thereby, contribute to more efficient selection of hybrid candidates.

Keywords: BLUP; GenPred; Genomic Selection; Shared Data Resources; genomic prediction; genomic selection; hybrid performance; maize; omics.

Figure 1

PC analysis of Dent (red) and Flint lines (teal) for G, T, S, and R data. The variances explained by PC 1 (x-axis) and PC 2 (y-axis) are shown in the respective captions.

Figure 2

Associations among off-diagonal elements of the kernel matrices for various predictors. Diagonal boxes: Densities of pairwise kernel coefficients among Dent (red) and among Flint (teal) parent lines. Off-diagonal boxes: Scatterplots of kernel coefficients for P, G, T, S, and R data with the Pearson correlation coefficients for pairwise comparisons between kernel matrices and labels defining the respective pair of predictors following the pattern “y-axis|x-axis.”

Figure 3

Predictive abilities (r) for T0 hybrids of single predictors (P, G, T, S, and R) and combinations thereof for GY and GDMC from 1000 CV runs with median r given above each column.

Figure 4

Predictive abilities (r) for T0 hybrids in GY and GDMC, respectively, for 66 cases that differ in their weights for the predictors P, G, and T. Their corresponding kernels were joined with weights varying from 0 to 1 in increments of 0.1. Weights for P (w_P) and G (w_G) are shown at the respective scales; weights for T are w_T = 1−w_P−w_G. Plotted values represent medians of r across 1000 CV runs. Heat color schemes differ for GY and GDMC, ranging from purple, indicating the respective lowest value, to yellow for the respective highest value.