- Genomic selection is a modern genetic improvement method that uses information from genome-wide genetic markers to predict the genetic merit of animals and identify the best candidates for breeding. It has transformed animal breeding by enabling more accurate selection at an early age, reducing dependence on lengthy progeny testing, and accelerating genetic progress in livestock populations. Genomic selection is widely used or investigated in dairy cattle, beef cattle, pigs, poultry, sheep, goats, and other breeding populations where genetic improvement is an important objective.
- Traditional selection methods rely on an animal’s observable performance, pedigree information, and estimated breeding values derived from relatives or offspring. Although these methods remain important, they may be less effective when traits are difficult or expensive to measure, expressed late in life, limited to one sex, or recorded only after slaughter. Genomic selection addresses some of these limitations by incorporating information from thousands or millions of genetic markers distributed throughout the genome. These markers help capture the effects of many genetic regions associated with economically important traits.
- The central concept of genomic selection is the genomic estimated breeding value (GEBV). A GEBV predicts an animal’s genetic merit using genome-wide marker information and a statistical model developed from a suitable reference population. The reference population consists of animals with both genomic data and reliable phenotypic records, or other appropriate genetic evaluation information. The relationships between their marker patterns and measured traits are used to estimate marker effects or genomic breeding values. These estimates can then be applied to genotyped animals whose genetic merit needs to be predicted.
- A crucial component of genomic selection is the reference population. Its size, genetic diversity, phenotype quality, and relationship to the animals being evaluated strongly influence prediction accuracy. A large reference population containing animals that are genetically similar to the selection candidates often improves prediction, although the benefit depends on the trait and the quality of the data. Reference populations may include information from multiple herds, generations, or related breeds when the statistical model and data structure support such integration. Prediction accuracy may decline when genomic predictions are transferred to populations that differ substantially from the original reference group.
- Genomic selection begins with the collection of reliable phenotypic, pedigree, and genomic information. Animals are genotyped using SNP genotyping arrays or other suitable genomic technologies, and their genetic marker data undergo quality control to identify missing, unreliable, or inconsistent genotypes. Phenotypic records are checked for accuracy, and relevant environmental and management effects are considered during genetic evaluation. Depending on the breeding program, the analysis may also include pedigree relationships, previous estimated breeding values, health records, reproduction data, and information about production systems.
- Statistical models are used to connect genomic information with genetic merit. One widely used method is genomic best linear unbiased prediction (GBLUP), which uses a genomic relationship matrix to estimate the genetic values of animals. Other approaches include Bayesian methods that estimate marker effects under different assumptions about the distribution of genetic effects. More advanced methods may incorporate haplotypes, sequence variants, or machine-learning techniques when these provide reliable predictive value. The appropriate method depends on the trait, available data, population structure, and the objectives of the genetic evaluation program.
- Genomic selection is particularly valuable for polygenic traits, which are influenced by many genetic variants, each often having a relatively small effect. Examples include milk yield, growth rate, feed efficiency, fertility, longevity, disease resistance, and carcass composition. Instead of depending only on a few significant markers identified through genome-wide association studies (GWAS), genomic selection combines information across the genome to predict overall genetic merit. This is important because many variants contributing to complex traits may not individually reach stringent statistical significance in association studies.
- One of the major advantages of genomic selection is that it can increase the accuracy of selection at an early age. Young animals can be genotyped and evaluated before they produce offspring or express traits that require many years to measure. In dairy cattle, for example, genomic evaluations can help identify promising bulls and heifers before extensive progeny testing is completed. Similar approaches can support earlier selection in other livestock species. Early and more informative selection can reduce generation intervals and improve the rate of genetic gain when integrated effectively into a breeding program.
- The expected rate of genetic improvement depends on several factors, including selection intensity, prediction accuracy, the amount of additive genetic variation, and the generation interval. A commonly used relationship is:
- ΔG/year = (i × r × σ_A) / L
- Here, ΔG/year represents expected genetic gain per year, i is selection intensity, r is the accuracy of selection, σ_A is the additive genetic standard deviation, and L is the generation interval. Genomic selection can improve annual genetic gain by increasing the accuracy of early selection and shortening the time between generations. The actual outcome depends on the breeding population, selection strategy, reproductive technology, and management of genetic diversity.
- Genomic selection can also improve breeding decisions for traits that are difficult, expensive, or impractical to measure directly on every selection candidate. These include feed efficiency, methane emissions, disease resistance, carcass traits, resilience, and some welfare-related characteristics. When suitable reference populations and reliable records are available, genomic predictions can provide information about genetic merit even when the trait itself has not been measured on the candidate animal. However, prediction quality depends on the strength of the relationship between the available data and the trait being evaluated.
- Another important application is multi-trait genomic selection, in which several traits are evaluated together. Many breeding objectives involve balancing production, fertility, health, longevity, adaptation, and animal welfare. Genetic correlations between traits mean that selecting for one characteristic may improve or reduce performance in another. Multi-trait evaluation combines information across traits to improve prediction where appropriate, while a selection index can combine estimated genetic merit with economic or strategic breeding weights. This allows breeders to make decisions that reflect the overall breeding objective rather than maximizing a single trait.
- Genomic selection can also support the management of genetic diversity. Because genome-wide markers provide information about genetic relationships, breeding programs can identify closely related animals, monitor genomic inbreeding, and design matings that balance genetic gain with the retention of variation. Genomic information may help avoid excessive use of a small number of highly ranked sires and reduce the accumulation of inbreeding. Strategies such as optimal contribution selection can help balance the genetic contribution of selected animals while pursuing long-term improvement. This is important because excessive inbreeding can increase the risk of inbreeding depression and reduce future selection potential.
- Despite its advantages, genomic selection has limitations. Developing a reliable reference population requires substantial investment in genotyping, phenotyping, data management, and statistical analysis. Prediction accuracy may be lower for traits with poor-quality records, low heritability, or limited representation in the reference population. Genomic predictions can also become less accurate across generations as genetic relationships and linkage patterns change, making regular evaluation and updating important. Differences between breeds, environments, and management systems may further affect the transferability of genomic predictions.
- Genomic selection does not eliminate the need for phenotypic recording or conventional genetic evaluation. High-quality performance records remain essential for training and updating prediction models, monitoring genetic trends, and detecting changes in trait expression. Environmental conditions, nutrition, health management, and production systems continue to influence observed performance. Genomic selection predicts genetic merit; it does not guarantee that an animal will achieve a particular performance level under every environment. Therefore, genomic predictions should be interpreted alongside phenotypic evidence and the intended production conditions.
- Responsible implementation also requires attention to genetic diversity, animal health, welfare, and long-term breeding objectives. Selecting animals solely on the basis of a high overall genomic score may unintentionally increase the frequency of undesirable traits if the breeding objective is incomplete. Breeding programs should monitor correlated responses, genetic defects, inbreeding, and the representation of different families or lines. Periodic validation of prediction accuracy and transparent reporting of the traits included in genomic evaluations help maintain confidence in the system.
- Overall, genomic selection is a powerful approach for improving the efficiency and accuracy of genetic evaluation in animal breeding. By combining genome-wide marker information with reliable phenotypic records and statistical prediction models, it enables earlier selection, shorter generation intervals, and potentially faster genetic progress. When integrated with clear breeding objectives, high-quality reference populations, and strategies for maintaining genetic diversity, genomic selection can contribute to productive, healthy, adaptable, and sustainable livestock populations.
