Prospects of using genomic selection in plant breeding

Author: Ahmad H. Sallam

** Postdoctoral research scientist, University of Minnesota.

** 495 Borlaug Hall, 1991 Upper Bufford Ave., St. Paul, MN, 55108.

sall0029@umn.edu

Abstract

Fusarium head blight (FHB) is a devastating disease that caused billions of dollars in losses to barley and wheat production in the United States. Deoxynivalenol (DON) is a potent mycotoxin present in fusarium infected grain. Excessive DON levels are a threat to both food safety and end-use quality. Therefore, FHB poses a challenge to barley and wheat breeders. No immunity has been identified in either barley and wheat germplasm, and the resistance is of a quantitative nature. Improvement of quantitative traits has relied mainly on phenotypic selection where the breeding value of each line is estimated from a limited number of phenotypic evaluations. Estimating the breeding value of a line using limited phenotypic evaluations is inaccurate for traits with low heritability. Traditional marker-assisted selection was proposed as an indirect selection method to select individuals based on their marker allele constitution. However, the use of traditional marker-assisted selection has been limited in complex traits because of its low power to detect QTL and bias in the estimated marker effects. Genomic selection can overcome the limitations of traditional marker-assisted selections by using genomewide marker-based predictions to improve and accelerate the breeding process. Genomic selection was found to be an effective breeding approach to reduce FHB severity and DON accumulation for both barley and wheat.

​

Keywords: marker-assisted selection, genomic selection, disease resistance