Mon, 05 Feb 2018
Genomic selection (GS) is a new method for improving economic traits in animal populations by utilising information from molecular markers, phenotypes and pedigree.
by Dinesh Thekkoot PhD
This method provides the opportunity to estimate the breeding values of selection candidates more accurately and much earlier in life and for traits that are difficult to measure like carcass and health traits. Because of this, it is considered as a paradigm shift for animal breeding and is being adopted by most of the livestock industries worldwide. In the current commercial animal breeding landscape, genomic information is mostly used to predict the breeding values of selection candidates. There are several other components in the breeding scheme that can be impacted by incorporating genomic information. These components interact with each other and hence need to be optimised to maximise the long-term genetic response. Some of these components specific to swine breeding are listed below
Phenotyping / genotyping strategies
The success of a GS programme depends on the size of the population of animals that have both genotypes and phenotypes (called as reference population), that are used to predict the genomic breeding values of the young selection candidates. To maintain the high prediction accuracies of genomic evaluation the reference population needs to be updated regularly. But, the number of phenotypes/genotypes that can be collected in a breeding program are limited and depends on (a) The cost of phenotyping/genotyping: because of the high cost it won’t be possible to collect phenotypes/genotypes on all animals and for all traits in the selection index. Also, the facilities required to measure accurate phenotypes might be limited. (b) The diminishing marginal returns as we allocate more resources for additional phenotypes /genotypes i.e. as the number of phenotype/genotype increases, the benefit of adding one extra phenotype/genotype decreases, even though the cost of collecting that phenotype/genotype remains the same. All these factors can influence the genetic gain in a breeding program, and hence should be optimised to maximise the long-term genetic gain.
Utilising cross bred information
Crossbreeding is widely used in the swine industry, and the genetic improvement goal is to improve crossbred performance under commercial conditions, while the selection is based on purebred performance in the nucleus/multiplication herds. Generally, the genetic correlation between the same trait recorded in the nucleus and commercial herds is lower than 1 indicating that the genetic control of the trait may differ between the nucleus and commercial herds (i.e. environments). For traits with a genetic correlation of 0.7 or less, it is suggested that data from commercial farms be incorporated into genetic evaluation programs to obtain greater genetic responses in the crossbred populations. Use of genomic information will help to incorporate the crossbred information in the routine evaluation, thereby increase the accuracies of estimated breeding values of crossbred performance from purebred data, and to maximise long-term genetic gain (Dekkers, 2007).
Pedigree mistakes can lower the rate of genetic gain. Genomic information can be used to confirm and correct the pedigree, and there by can maximise the long-term genetic gain.
Re-evaluate selection indexes
To maximise the long-term genetic gain, we can think of modifying the breeding objectives. With the availability of genomic information, we have the opportunity to modify the breeding objective by incorporating traits such as health and fitness along with appropriate economic values for these traits. Also, as genomic selection increases the rate of genetic gain, we need to re-evaluate and redefine the breeding objectives more often (Henryon et al., 2014).
Genomic information can be used to redesign selection methods to maximize long-term genetic gain by controlling the rates of inbreeding, to maintain sufficient genetic variation, to adjust the nucleus population size, increase the intensity of selection (details can be found here), etc.
Generally, mating designs receive less attention than other components while thinking of maximising long-term genetic gains. Any increase brought in by modifying a mating design can be considered as an added benefit to any breeding scheme as most of the changes in the mating scheme can be introduced without any extra cost or logical constraints. Genomic information will help us to modify the mating designs so as to disperse the genetic contribution more effectively by reducing the rate of inbreeding (Henryon et al., 2014).
Genesus has been conducting research in the field of genomic evaluation (GE) and selection from 2011 onwards, by investing in technology to accurately measure difficult traits, creating large populations of animals with genotypes and all phenotypes, simulating various GS scenarios, mimicking real population structures and data and validating the GE and GS programs. Recently, Genesus has incorporated GE and GS in their breeding programs to maximise the long-term genetic response and thereby to maximise the profitability of Genesus customers.
Dekkers, J. C. M. 2007. Marker-assisted selection for commercial crossbred performance. J. Anim. Sci. 85:2104–2114
Henryon M. et al., 2014. Animal breeding schemes using genomic information need breeding plans designed to maximize long-term genetic gains. J. Livsci. 166:38-47