|DUNNE, JEFFREY - North Carolina State University|
|SAMAYOA, LUIS FERNANDO - North Carolina State University|
|Holland, Jim - Jim|
Submitted to: Book Chapter
Publication Type: Book / Chapter
Publication Acceptance Date: 4/2/2018
Publication Date: N/A
Interpretive Summary: This book chapter reviews the principles of using genetic information for crop improvement. Some traits are strongly influenced by individual genes, in such cases, genetic mapping or association analysis followed by functional analysis may identify the causative gene and sequence variations. Rapid and inexpensive genetic tests for these variants can be used by crop breeders to efficiently select the favorable genes in breeding populations. For other traits, many genes each with very small effects control most of the genetic variation. In these cases, identifying the causative genes is very difficult and even if successful does not lead to useful genetic markers for breeding. An alternative approach is genomic selection, which exploits genetic information throughout the genome to select individuals that have the best combinations of many genetic variants. Genomic selection has proven effective and is most efficient when it can be employed in circumstances that prevent accurate trait measurements. Finally, some traits are influenced by a combination of a few genes of large effect plus many genes with small effects. In such cases, the two approaches can be combined to select for specific large genetic effect variants plus ‘genetic backgrounds’ characterized by genome-wide random markers.
Technical Abstract: The use of genetic information to predict the value of individuals in plant breeding populations began about 40 years ago. The original paradigm was to identify genomic regions with outsize influence on a trait of economic value, then to use markers in that genomic region to select individuals carrying the desired allelic variants. An explosion of interest in mapping such quantitative trait loci followed, with thousands of genomic regions associated with important traits across many species. The practical use of such information lagged well behind the discovery of quantitative trait loci, however, due mostly to the problem that individual markers were often only associated with a small proportion of genetic variation, such that their value in selection was very small. In a few lucky cases, individual genes with very large effects on important traits were discovered, and these could be more easily turned into useful selection targets. Genome-wide association studies have powered up the ability to identify individual variants associated with useful effects in crops, but the fundamental problem of accurately estimating marker effects and using them in selection remains for traits affected by many genes. Genomic selection was proposed by animal breeders as a way to more effectively use the information contained in dense genetic marker sets for the prediction of quantitative traits. Crop breeders subsequently discovered that this approach could be generalized across the diverse population structures and mating systems of plants and have begun implementing genomic selection in crops with success.