Poster

Harika Pothula ¹, Evan Salsman ¹, Jason Axtman ¹, Justin Hegstad ¹, Yueqiang Leng ², Zhao Jin ¹, Shaobin Zhong ³, Jason Fiedler ⁴, Steven Xu ⁵, Elias Elias ¹, and Xuehui Li ¹
1. Department of Plant Sciences, North Dakota State University, Fargo, ND, USA
2. Department of Plant Pathology, North Dakota State University, Fargo, ND, USA
3. USDA-ARS Cereal Disease Laboratory, St. Paul, MN, USA
4. USDA-ARS Edward T. Schafer Agricultural Research Center, Fargo, ND, USA
5. USDA-ARS Crop Improvement and Genetics Research, Western Regional Research Center, Albany, CA, USA
Corresponding author: Xuehui Li, xuehui.li@ndsu.edu

Pothula, Harika

harika.pothula@ndsu.edu

Durum wheat (Triticum turgidum L. subsp. durum) is highly susceptible to Fusarium head blight (FHB), which reduces grain yield and quality. Breeding for resistance is challenging because resistant germplasm is limited and resistance is quantitatively inherited. We evaluated recurrent phenotypic selection (PS) and recurrent genomic selection (GS) as complementary approaches to improving FHB resistance while maintaining acceptable plant height and flowering time. A base population was developed by crossing ten moderately resistant lines, derived from tetraploid wheat relatives and hexaploid bread wheat, with five elite durum lines and cultivars. Five annual cycles of PS for FHB severity were completed from 2019 to 2024. Mean FHB severity declined from 71.2% in Cycle 0 to 52.1% in Cycle 5, while plant height and flowering time remained within breeding targets. Genomic prediction models were trained on 318 NDSU durum breeding lines and 438 S_0:1 lines from PS Cycles 3 to 5. Prediction accuracies were 0.53 for FHB severity, 0.53 for plant height, and 0.79 for days to flowering. Two GS cycles were completed between September 2024 and April 2025, starting from the 15 highest-ranked S_0:1 lines of PS Cycle 5. In 2025 field trials at Fargo and Prosper, North Dakota, the GS Cycle 1 and Cycle 2 populations had lower mean FHB severity (42.8% and 41.8%, respectively) than the PS Cycle 5 population. Selected lines also accumulated less deoxynivalenol (DON): 16 of 24 lines at Fargo and 8 of 20 at Prosper had significantly lower DON than ND Riveland (P < 0.05). Together, these results show that recurrent PS and GS can improve FHB resistance in durum wheat and generate promising germplasm for cultivar development in the Northern Great Plains. Keywords: durum wheat; Fusarium head blight; recurrent selection; genomic prediction; deoxynivalenol

© 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).