USWBSI

USWBSI Abstract Viewer

2026 National Fusarium Head Blight Forum


Variety Development and Host Resistance (VDHR)

Poster # 517

Evaluating Family-Based Genomic Prediction Strategies for Fusarium Head Blight Traits in SSRWW

Authors & Affiliations:

Julio Sellani 1, Jeanette Lyerly 3, Stephen Harrison 1, Trey Paul Price 1, Richard Boyles 2, Noah DeWitt 1
1. Louisiana State University, School of Plant, Environmental and Soil Sciences, Baton Rouge, LA,
2. Clemson University, Department of Plant and Environmental Sciences, Florence, SC,
3. North Carolina State University, Department of Crop and Soil Sciences, Raleigh, NC
Corresponding Author: Noah DeWitt, NDeWitt@agcenter.lsu.edu

Presenting Author:

Sellani, Julio
jsellani@agcenter.lsu.edu

Abstract:

Genomic selection (GS) can accelerate genetic gain for Fusarium head blight (FHB) resistance in wheat by enabling earlier selection. However, individual genotyping costs may limit its implementation in early-generation breeding populations. This study evaluated whether family-level genomic information could achieve predictive ability comparable to individual genotyping while potentially reducing genotyping requirements. A soft red winter wheat panel comprising approximately 450 lines from 90 full-sib families was evaluated under artificial FHB inoculation across Louisiana environments. Traits included FHB visual severity, Fusarium-damaged kernels (FDK), deoxynivalenol (DON), heading date (HD), and stripe rust (yellow rust; YR). Genomic best linear unbiased prediction (GBLUP) was implemented using three methods for assigning marker calls to individual genotypes: individual genotypes (IND), family-mean marker scores (FM), and imputed mid-parent genotypes (MP). Predictive ability was assessed through leave-one-family-out (LOFO) five-fold cross-validation. Genome-wide association studies (GWAS) and forward validation using historical genomic predictions were also conducted.Predictive ability varies across traits and genomic representations. Across combined evaluations, predictive abilities for IND, FM, and MP were 0.32, 0.33, and 0.32 for FHB visual severity; 0.41, 0.42, and 0.43 for FDK; and 0.31, 0.32, and 0.33 for heading date, respectively. DON showed lower predictive ability, ranging from 0.22 to 0.27. Family-based approaches generally performed comparably to individual genotyping. GWAS identified significant associations for FDK and heading date. Forward validation using historical breeding data also demonstrated comparable predictive abilities, suggesting that historical genomic information can support family-level prediction in new breeding populations.These findings suggest that family-level genomic information can provide useful predictive ability for FHB-related traits. Prediction of families using mid-parent marker scores from sequenced parents, may offer comparable predictive ability to mass genotyping of progeny. Integrating family-based genomic prediction with individual-level evaluation at later breeding stages could improve selection efficiency and support the development of FHB-resistant wheat cultivars.


© 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).