Genomewide association and prediction of phenotypic stability in barley

Authors: Neyhart, Jeffrey; GUTIERREZ, LUCIA - University Of Wisconsin; SMITH, KEVIN - University Of Minnesota

Submitted to: The Plant Genome
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/11/2025
Publication Date: N/A
Citation: N/A

Interpretive Summary: Climate change threatens crop production through an increase in the occurrence of abiotic stresses such as drought and extreme temperatures. Breeding new crop varieties that are more tolerant to these extremes may be accomplished by selecting for stability of yield, quality, or other traits across environments; however, measuring stability requires data, which is a resource intensive process involving many field experiments under different growing conditions”. To address this problem, we used data from an experiment in barley in which populations were jointly phenotyped in 42 different environments for five agronomic and malting quality traits. We calculated trait stability for each barley variety across environments and used molecular markers to identify 87 significant associations between markers and trait stability. We found extensive overlap between genomic regions associated with trait stability, indicating that genes influencing overall trait variation also influence trait stability across environments. We also discovered that genomewide prediction accuracy was moderate-to-high when attempting to predict the stability, even when using only a subset of environments to calculate stability. These predictions will benefit plant breeders in selecting for more stable varieties without resource-intensive multi-environment trials.

Technical Abstract: Climate change threatens crop production through an increase in the occurrence of abiotic stresses such as drought and extreme temperatures. Breeding and growing cultivars that are more genetically tolerant of these stresses, or those with greater phenotypic stability, may help to mitigate the impact of environmental stresses. Understanding the genetic architecture and marker-based predictive ability of stability could increase selection progress. Using a barley (Hordeum vulgare L.) multi-environment dataset, our objectives were to (a) identify genomic regions associated with per se means and stability for 5 agronomic and malting quality traits; (b) determine the genomewide prediction accuracy of stability; and (c) assess the impact of subsampling environments on estimates of stability. Data on a 233-line founder and offspring population grown in 42 environments and phenotyped for five traits was used to calculate the per se means and stability (both linear and non-linear) for each line for each trait. We identified 87 marker-trait associations, and nearly all significant SNPs for linear stability overlapped with previously discovered trait mean QTL in barley, suggesting shared genetic control. Genomewide prediction accuracy of linear stability was moderate as measured using cross-validation (rMP = 0.32 - 0.69) and remained so when making predictions of an unobserved offspring test population (rMP = 0.26 - 0.61). Increasing the number of randomly sampled environments from which to draw phenotypic data led to more precise estimates of stability, greater marker-trait discovery rates, and higher genomewide prediction accuracy; however, a modest number of environments was sufficient for obtaining reliable predictions. Results indicate that the genetic control of stability and per se means is likely shared, but predictive breeding can aid in making selection decisions in the absence of resource-intensive multi-environment trials.