- Genomic prediction of fertility is the process of estimating an animal’s genetic potential for reproductive performance using DNA marker information combined with performance records, pedigree information, and statistical models. It is an important component of modern genomic selection because fertility traits directly influence reproductive efficiency, production costs, offspring survival, and the long-term profitability of livestock breeding programs. Genomic prediction can help breeders identify animals with favourable fertility genetics at an early age, even before they have produced offspring or accumulated extensive reproductive records.
- Fertility is a complex quantitative trait influenced by many genes, environmental conditions, and management practices. Genetic factors may affect age at puberty, estrous cycles, ovulation, semen quality, fertilization, embryo survival, pregnancy maintenance, and successful delivery. Because these traits are often difficult or expensive to measure and may have low or moderate heritability, genomic prediction offers an opportunity to improve the accuracy of genetic evaluation by using information distributed across the genome.
- The process begins with collecting reliable phenotypic records, such as conception rate, pregnancy rate, services per conception, calving interval, age at first calving, litter size, and semen quality. These records are linked to the animals’ DNA marker profiles, commonly obtained using single-nucleotide polymorphism (SNP) genotyping. Statistical models identify associations between genome-wide marker patterns and observed fertility performance. The resulting predictions estimate the genetic merit of animals for specific reproductive traits rather than simply describing their current reproductive condition.
- A genomic prediction model is developed using a reference population of animals with both genomic information and relevant performance records, or reliable genetic evaluations derived from those records. The model estimates how marker patterns relate to genetic differences in fertility. Once trained and appropriately validated, it can predict the genomic breeding values of younger animals that have not yet expressed all their reproductive traits. The accuracy of prediction depends on factors such as the size and genetic composition of the reference population, the quality of the phenotypic records, trait heritability, marker density, and the relationship between reference and candidate animals.
- Genomic estimated breeding values (GEBVs) are used to rank animals according to their predicted genetic merit. Animals with favourable GEBVs for fertility may be selected as breeding candidates, allowing genetic improvement to begin earlier and potentially shortening the generation interval. This is especially useful for traits that are expressed only in one sex, require maturity to measure, or depend on repeated reproductive records. However, a favourable GEBV does not guarantee successful conception or pregnancy because nutrition, health, age, housing, heat stress, and other environmental conditions remain important.
- Genomic prediction can be applied to both female and male fertility. In females, it may support selection for earlier reproductive maturity, improved conception, reduced pregnancy loss, appropriate litter size, and better reproductive longevity. In males, genomic predictions may complement information on semen quality, libido, and reproductive performance. The usefulness of each prediction depends on the accuracy of the underlying data and the relevance of the model to the species, breed, and production system.
- Different statistical approaches can be used for genomic prediction, including genomic best linear unbiased prediction (GBLUP) and Bayesian methods that estimate genetic effects from genome-wide markers. These methods account for genetic relationships and marker information in different ways. More advanced approaches may also incorporate biological knowledge or machine-learning techniques, but their performance must be tested using independent data rather than assumed to be superior. Proper validation helps determine whether predictions remain reliable for animals outside the original reference population.
- One important challenge is that fertility traits are often recorded inconsistently across farms and breeding systems. Differences in management, disease exposure, nutrition, and reproductive protocols can influence recorded performance and reduce the reliability of genetic predictions if they are not adequately accounted for. Genotype–environment interaction may also occur when animals with favourable genetic merit in one production environment do not perform equally well in another. Maintaining accurate records, using appropriate statistical adjustments, and updating reference populations can improve prediction reliability over time.
- Genomic prediction should also consider genetic relationships between fertility and other economically important traits. Selection for rapid growth or high production may sometimes be associated with reduced reproductive performance, depending on the population and breeding objective. Selection indexes can combine genomic predictions for fertility, production, health, longevity, and welfare to support balanced decisions. Breeders should also monitor inbreeding and genetic diversity to reduce the risk of concentrating harmful alleles within a population.
- Overall, genomic prediction of fertility provides a powerful tool for improving reproductive performance and accelerating genetic progress in livestock. Its success depends on reliable reproductive phenotypes, representative reference populations, validated statistical models, and responsible breeding decisions. When combined with conventional genetic evaluation, good management, and long-term monitoring, genomic prediction can help identify animals with favourable reproductive genetics while supporting sustainable and efficient animal production. Related topics include genetic selection for fertility, genetic control of fertility, breeding values, heritability, genetic correlation, genomic selection, embryonic survival, and reproductive disorders and genetics.
