Poster

Rafael Bley Cottica ¹, Stephen Wegulo ², and Katherine Frels ¹
1. Department of Agronomy and Horticulture, University of Nebraska, Lincoln, NE 68583, USA
2. Department of Plant Pathology, University of Nebraska, Lincoln, NE 68583, USA
Corresponding Author: Katherine Frels; Email: kfrels2@unl.edu

Bley Cottica, Rafael

rbleycottica2@huskers.unl.edu

The University of Nebraska Small Grains Breeding Program has operated a unique winter barley breeding program for 50 years, with doctoral students serving as breeders under the principal investigator's supervision. Frequent breeder turnover can slow progress; historical data from 2002 to 2022 showed a genetic gain of 0.05 bushel per acre per year. Genomic selection (GS) can standardize parts of selection and advancement, and a Bayes A model previously predicted yield in the program with R2 = 0.41. Fusarium head blight (FHB) reduces grain weight and quality and causes accumulation of deoxynivalenol (DON), leading to market rejection. This study evaluated GS for FHB resistance using 152 breeding lines genotyped with the USDA-SoyWheOatBar-3K array (1,850 SNPs after quality control) and FHB nursery data from 2022 to 2026. Severity, incidence, FHB index, Fusarium-damaged kernels (FDK), and DON were predicted with a GBLUP model that included a genotype × year interaction, using leave-one-year-out validation. For new, unphenotyped lines, predictive ability averaged 0.24 for severity and 0.28 for FHB index, but was near zero for FDK and DON. When half of a season's lines were phenotyped (sparse testing), predictive ability for the remaining lines rose to 0.33 for severity, 0.34 for FHB index, and 0.21 for DON, and was more consistent across years for severity and FHB index. The genotype × year interaction was substantial for severity, FHB index, and DON (between-year genetic correlation ≈ 0.66). GBLUP and BayesB performed equally, indicating that FHB resistance in this germplasm is polygenic. GS combined with sparse field testing is a promising tool for selecting FHB-resistant lines, while DON prediction still requires partial phenotyping within each season.

© Copyright 2026 by individual authors. All rights reserved. No part of this abstract or paper publication may be reproduced without prior permission from the applicable author(s).