Managing Global Resources for a Secure Future

2017 Annual Meeting | Oct. 22-25 | Tampa, FL

336-2 High-Throughput Phenotyping for Plant Characterization, Selection, and Breeding.

See more from this Division: C08 Plant Genetic Resources
See more from this Session: Symposium--Phenotyping Plant Genetic Resources to Support Climate Smart Agriculture

Wednesday, October 25, 2017: 8:33 AM
Tampa Convention Center, Room 1

Arron Carter1, Jayfred Gaham Godoy1, Stephanie Sjoberg1 and Kimberly A. Garland Campbell2, (1)Crop and Soil Sciences, Washington State University, Pullman, WA
(2)Wheat Genetics, USDA-ARS Washington State University, Pullman, WA
Abstract:
Recent advances in genomics and phenomics have opened the path for many new strategies in plant characterization and breeding. These tools can provide better understanding and prediction of complex traits necessary to create varieties with superior agronomic performance and tolerance to abiotic and biotic stresses. Genomics and high-throughput phenotyping offers the advantage of making gains using indirect selection methods in breeding programs, as well as characterizing germplasm in large collections. Two large diversity panels of wheat were developed and grown over four years in the Pacific Northwest. Data was collected on both populations for agronomic traits, yield potential, canopy spectral reflectance and disease resistance. Prediction models using spectral reflectance indices calibrated on one panel were used to make selections on the second panel. Vegetation (NDVI) and water (NWI3) indices taken at early grain-filling stage consistently discriminated low and high yielding genotypes and at the same time effectively selecting the best performing lines across years, locations and environmental conditions (dry and irrigated). The ability to select for these traits using indirect selection methods may advance the rate of gain for these correlated traits. Using a limited number of molecular markers associated with traits for abiotic stress tolerance, increased gains were not achieved. Whole genome approaches like genomic selection were able to select lines with improved performance. Genomic selection using genomic best linear unbiased prediction (gBLUP) models gave the best predictive accuracies (~44-51%) for yield in both panels when compared to other genomic prediction models. gBLUP showed higher prediction accuracy for winter survival (43-53%), aluminum tolerance (44%) and resistance Cephalosporium stripe (24%). Whole genome or whole plant approaches can be applied during early generation selection in plant breeding. Indirect selection using high-throughput phenotyping alone or in combination with genomic selection can expedite development of high yielding and broadly adapted wheat varieties.

See more from this Division: C08 Plant Genetic Resources
See more from this Session: Symposium--Phenotyping Plant Genetic Resources to Support Climate Smart Agriculture