Poster # 509

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

Xuehui Li

Fusarium Head Blight (FHB) is a major disease that can cause severe loss of grain yield and quality of durum wheat in the Northern Great Plain of US. Developing FHB resistant germplasm and cultivars is the key to minimize the loss. Hundreds of QTL have been identified for FHB resistance in wheat. A few of resistant alleles were found in elite durum wheat lines, but majority were found in bread wheat and wild relatives. Recurrent selection is an effective way to increase frequencies of resistant alleles and to develop pre-breeding germplasm with improved FHB resistance. A base population was developed using 15 founders including five elite durum wheat breeding lines and 10 FHB resistant durum lines, where resistant alleles were introgressed from wild tetraploid wheat relatives or bread wheat. Four cycles of recurrent phenotypic selection were conducted from 2019 to 2023, one cycle per year. The FHB index was reduced from 82.3% in Cycle 0 to 48.5% in Cycle 4, proving that recurrent phenotypic selection can improve FHB resistance in the durum wheat population. Genomic selection (GS) can speed up selection and increase genetic gain in term of time and cost. We evaluated genomic prediction for FHB index using 318 elite durum wheat breeding lines (including old cultivars, modern cultivars, and breeding lines from the NDSU durum wheat program) and 256 S_0:1 lines from the recurrent selection Cycle 3 and Cycle 4 populations. The total 574 lines were genotyped using 40K SNP array, and a total of 4,048 SNP markers were obtained with missing values less than 10% and minor allele frequency greater than 5%. The marker data indicated little decrease in genetic diversity from the 15 founders to Cycle 4 population. The 318 elite durum wheat lines were evaluated in the FHB field nurseries in 2019, 2021, and 2023. Using rrBLUP model with cross-validation, the prediction accuracies for FHB index, plant height, and days-to-flowering were 0.51, 0.53, 0.79, respectively. Using the 318 elite breeding lines and 122 S_0:1 lines from the Cycle 3 population as training population to predict the Cycle 4 population, prediction accuracies were 0.27, 0.39,and 0.63, respectively. Our results indicated that implementing GS in the recurrent selection can potentially accelerate the genetic improvement.