27-2 Applying Genomic Selection to Crops In the Public Sector.

See more from this Division: C01 Crop Breeding & Genetics
See more from this Session: Symposium--Genomic Selection Based on High Density Marker Data
Sunday, October 16, 2011: 3:30 PM
Henry Gonzalez Convention Center, Room 214D
Share |

Jean-Luc Jannink, 407 Bradfield Hall, USDA-ARS, Ithaca, NY
Genomic selection (GS) uses a training population that has been phenotyped and genotyped to develop a model to predict breeding values of individuals that have only been genotyped. Individuals can be selected on these predictions prior to phenotyping, thereby accelerating the breeding cycle. GS models are predictive, not mechanistic though they can incorporate knowledge from mechanistic studies. The goal of GS model development is really to evaluate the main effect of alleles over environments and genetic backgrounds to provide a stable selection criterion. Thus experimental breeding methods need to shift their focus from evaluating lines to evaluating alleles. Breeding programs can increase response in several ways: increasing the selection differential by reducing the fraction selected; increasing the accuracy of the selection criterion by replicating more; increasing the genetic variance by introgressing favorable alleles; or increasing the cumulative selection differential by accelerating the breeding cycle. Only the last has a linear return on effort while for others return is less-than-linear. Thus the ability of GS to accelerate the breeding cycle is the key to its efficiency. To initiate GS, public sector breeders must think about the training population and how to allocate costs between genotyping and phenotyping. Furthermore, breeding schemes need to account for whether selection candidates will be inbred or not, and how many crossed seeds it is possible to obtain from a cross. In this talk, I will introduce GS and illustrate it with a few possible breeding schemes to make it a more concrete possibility.
See more from this Division: C01 Crop Breeding & Genetics
See more from this Session: Symposium--Genomic Selection Based on High Density Marker Data