Genomic Selection for Crop Improvement

Authors: HEFFNER, ELLIOT - CORNELL UNIVERSITY; SORRELLS, MARK - CORNELL UNIVERSITY; Jannink, Jean-Luc

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/13/2008
Publication Date: 1/28/2009
Citation: Heffner, E.L., Sorrells, M.E., Jannink, J. 2009. Genomic Selection for Crop Improvement. Crop Science. 49:1-12.

Interpretive Summary: Despite important strides in marker technologies, the use of marker-assisted selection (MAS) has stagnated for the improvement of quantitative traits. Bi-parental mating designs for the detection of loci affecting these traits (QTL) impede their application to breeding, and the statistical methods used do not identify small-effect QTL that are common for these traits. Genomic selection (GS) has been proposed to address these deficiencies. Genomic selection analyzes phenotypes and high-density marker scores of lines in a breeding population to predict the value of lines as parents. To improve prediction, GS incorporates all marker information in the prediction model, thereby avoiding biased marker estimates and capturing more of the variation due to small-effect QTL. In simulations, the correlation between true performance and the GS-predicted performance reaches levels of 0.85 even for complex low heritability traits. This level of prediction accuracy is sufficient to consider selecting for agronomic performance using marker information alone. Selection on markers alone would substantially accelerate the breeding cycle, enhancing gains per unit time. It would dramatically change the role of phenotyping, which would then serve to update prediction models and no longer to select lines. While research to date shows the promise of GS, work remains to be done to validate it empirically, to determine where it is best suited, and to incorporate it into breeding schemes.

Technical Abstract: The use of marker-assisted selection (MAS) has stagnated for the improvement of quantitative traits. Experimental designs for the detection of loci affecting these traits impede their application, and the statistical methods used are ill-suited to the traits’ polygenic nature. Genomic selection (GS) has been proposed to address these deficiencies. Genomic selection encompasses methods to predict breeding values by analyzing phenotypes and high-density marker scores of lines in a breeding population. A key to the success of GS is that it incorporates all marker information in the prediction model, thereby avoiding bias associated with selecting model variables and capturing more of the variation due to small effect QTL. In simulations, the correlation between true breeding value and the genomic estimated breeding value from GS reaches levels of 0.85 even for highly polygenic low heritability traits. This level of prediction accuracy is sufficient to consider selecting for agronomic performance using marker information alone. Selection on markers alone would substantially accelerate the breeding cycle, enhancing gains per unit time. It would dramatically change the role of phenotyping, which would then serve to update prediction models and no longer to select lines. While research to date shows the promise of GS, work remains to be done to validate it empirically, to determine where it is best suited, and to incorporate it into breeding schemes.