A recent study conducted by several researchers from Animal Breeding and Genomics (ABG) and the Laboratory of Genetics of Wageningen University & Research and INRA in France investigated the long-term effects of genomic selection. They show that genomic selection outperforms pedigree selection in terms of genetic gain, but results in a similar reduction of genetic variance. By performing a series of simulations, the researchers were able to map the changes in the genetic architecture of traits as a result of selection and, consequently, show that the genetic architecture of traits changed considerably across generations, especially when selection was accurate.
From pedigree to genomic selection
Recently, genomic selection has revolutionized animal breeding and is currently used in most of the breeding programmes around the world; in some breeding programs genomic selection has doubled the annual rate of genetic gain compared to pedigree-based selection. This is mainly because genomic selection enables accurate selection of animals at a young age. Moreover, genomic selection enables selection for lowly-heritable traits and for traits that are difficult to measure, for which pedigree-based selection is generally not very effective. Within genomic selection, genotypes on several thousands of DNA markers covering the genome, along with recorded phenotypes, are used to identify the genetically best animals.
However, very little is known about the long-term effects of genomic selection, and whether genomic selection is able to maintain genetic gain in the future. Consequently, the authors designed a study to examine the long-term effects of genomic selection on response to selection, genetic variance, and the genetic architecture of traits using stochastic simulations.
Simulating the genetic architecture of a trait
The authors defined the genetic architecture as the set of causal loci (the positions on a chromosome with an effect on the trait) underlying each trait, their allele frequencies, and their statistical additive effects. To investigate the long-term effects of genomic selection on the genetic architecture of traits, they compared genomic selection to phenotypic and pedigree-based selection. They simulated a livestock population under 50 generations of selection for a single trait, controlled by either only additive, additive and dominance, or additive, dominance and epistatic effects.
The effects of (genomic) selection on the genetic architecture of a trait
The study shows that short-term response to selection was always greatest with genomic selection. On the long-term, genomic selection always outcompeted pedigree selection in terms of cumulative response to selection. However, phenotypic selection has the potential to obtain a higher genetic gain than genomic selection over many generations of selection. The reason is that both genomic selection and pedigree selection lose a considerable amount of genetic variance, which was much larger than the loss with phenotypic selection.
Pedigree and genomic selection lost a similar amount of genetic variance after 50 generations of selection, but genomic selection maintained more segregating loci, which on average had lower minor allele frequencies than with pedigree selection. Based on this result, genomic selection is expected to better maintain genetic gain after 50 generations than pedigree selection.
The genetic architecture of traits is known to evolve over time. As a result of selection, the genetic architecture is changing faster, because the subset of loci and their allele frequencies are changing. This can also change the allele substitution effects of the loci over time when non-additive effects (such as dominance and epistasis) are present. Those changes reduce the informativeness of previous generations for genomic prediction.
Long-term genetic gain
The authors conclude that their research shows that genomic selection outperforms pedigree selection in terms of long-term genetic gain, but results in a similar reduction of genetic variance. They state that “the genetic architecture of traits changed considerably across generations, especially under selection and when non-additive effects were present.” In conclusion, non-additive effects had a significant impact on the accuracy of selection and long-term response to selection, particularly when selection was accurate.
The study “The long-term effects of genomic selection: 1. Response to selection, additive genetic variance, and genetic architecture” is part of the project ‘(R)evolution of traits? Quantifying the genetic change in traits over generations as a result of Genomic Selection’ of the research programme Veni, which is (partly) financed by the Dutch Research Council (NWO).
Preparations for a follow-up study are already in operation. The second part of the study will focus on uncovering more information about the genes that contribute to genetic improvement; how much of an effect do these genes actually have? What happens to genes that have a high frequency? And what about genes that have a low frequency? By looking at the changes that result from selection, the researchers aim to find out if the nature of these changes can be linked to the specific method of selection that was used.