Multi-environment and multi-trait genomic selection models in unbalanced early generation wheat yield trials

Authors: WARD, BRIAN - North Carolina State University; Brown-Guedira, Gina; KOLB, FREDERIC - University Of Illinois; VAN SANFORD, DAVID - University Of Kentucky; TYAGI, PRYANKA - North Carolina State University; SNELLER, CLAY - The Ohio State University; GRIFFEY, CARL - Virginia Polytechnic Institution & State University

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/20/2019
Publication Date: 4/4/2019
Citation: Ward, B.P., Brown Guedira, G.L., Kolb, F.L., Van Sanford, D.A., Tyagi, P., Sneller, C.H., Griffey, C.A. 2019. Multi-environment and multi-trait genomic selection models in unbalanced early generation wheat yield trials. Crop Science. 59:491-507.

Interpretive Summary: The majority of studies evaluating genomic selection (GS) for plant breeding have utilized single-trait, single-site models which ignore genotype-by-environment interaction (GEI) effects. However, such studies do not accurately reflect the complexities of many applied breeding programs where breeders need to select for multiple traits and many environments. Previous papers have found that models that incorporate GEI effects and multiple traits can increase the accuracy of genomic prediction. This study’s goal was to test GS methods for prediction in scenarios that simulate early-generation yield testing by correcting for field spatial variation and fitting multi-environment and multi-trait models on data for 14 traits of evaluated in soft red winter wheat. Corrections for spatial variation across fields increased across-environment trait heritability by 25% on average but had little effect on model predictive ability. Results between all models were generally equivalent when predicting the performance of newly introduced breeding lines. However, models incorporating GEI information and multiple traits increased prediction accuracy by up to 9.6% for low-heritability traits when phenotypic data was sparsely collected across environments. The results suggest that GS models utilizing multiple traits and incorporating GEI effects may best be suited to predicting line performance in new environments when phenotypic data has already been collected across a subset of the total testing environments.

Technical Abstract: The majority of studies evaluating genomic selection (GS) for plant breeding have utilized single-trait, single-site models which ignore genotype-by-environment interaction (GEI) effects. However, such studies do not accurately reflect the complexities of many applied breeding programs, and previous papers have found that models that incorporate GEI effects and multiple traits can increase the accuracy of genomic estimated breeding values (GEBVs). This study’s goal was to test GS methods for prediction in scenarios that simulate early-generation yield testing by correcting for field spatial variation, and fitting multi-environment and multi-trait models on data for 14 traits of varying heritability evaluated in unbalanced designs across four environments. Corrections for spatial variation increased across-environment trait heritability by 25% on average, but had little effect on model predictive ability. Results between all models were generally equivalent when predicting the performance of newly introduced genotypes. However, models incorporating GEI information and multiple traits increased prediction accuracy by up to 9.6% for low-heritability traits when phenotypic data was sparsely collected across environments. The results suggest that GS models utilizing multiple traits and incorporating GEI effects may best be suited to predicting line performance in new environments when phenotypic data has already been collected across a subset of the total testing environments.