# Dairy Cattle Genomics: Breeding Values, Genomic Testing, and Selection ## Key Takeaways - Genomic testing in dairy cattle utilizes DNA samples to predict an animal's genetic merit at birth, providing Genomic Predicted Breeding Values (gEBVs) and composite selection indexes that rank animals for profitability or efficiency, significantly shortening the generation interval compared to traditional progeny testing. - gEBVs are calculated by statistically linking thousands of DNA markers to extensive phenotypic performance data from a reference population, with accuracy influenced by the reference population's size and relevance, trait heritability, and genetic distance. - The practical application involves ranking young animals at birth to inform decisions on heifer retention, calf sales, and sire selection, with composite indexes serving as primary ranking tools, supplemented by individual trait gEBVs to address specific herd needs. - Accurate genomic selection is critically dependent on high-quality phenotypic records (milk yield, somatic cell count, fertility, conformation, etc.) submitted by farmers to national databases, with accurate animal identification being paramount for linking genetic data to performance. - Common failure patterns include testing without a clear selection plan, over-reliance on predictions for low-heritability traits, ignoring the genetic base year, and testing animals genetically distant from the reference population, all of which reduce the effectiveness of genomic selection. - Genomic selection can enhance animal welfare by enabling selection for health and fertility traits, but a sole focus on production can negatively impact metabolic health and longevity; biosecurity during sample collection and worker safety are also critical considerations. --- ## At a Glance Genomic testing in dairy cattle uses a DNA sample to predict the genetic merit of an animal at birth, long before traditional progeny testing can produce reliable results. The core output is a set of Genomic Predicted Breeding Values (gEBVs) and a composite selection index that ranks animals for profitability or efficiency. The table below summarizes the main genomic testing options available to dairy farmers and breeders. | Testing Option | Typical Sample | Primary Output | Common Use Case | |----------------|----------------|----------------|-----------------| | Low-density SNP chip (e.g., 9K to 50K markers) | Hair roots, blood, or tissue | gEBVs for production, health, and conformation traits | Screening heifer calves for retention or sale | | Medium-density SNP chip (e.g., 50K to 150K markers) | Hair roots, blood, or tissue | gEBVs with higher accuracy, parentage verification | Selecting replacement heifers and young sires | | High-density SNP chip (e.g., 150K+ markers) | Blood or tissue | gEBVs for research, fine-mapping, or rare trait prediction | Elite bull dams, research herds, or breed association requirements | Genomic testing does not replace traditional performance recording. It adds early-life prediction accuracy that shortens the generation interval and accelerates genetic gain. The practical outcome for a dairy farmer is a list of young animals ranked by predicted genetic merit, which can be used to decide which heifers to breed from, which calves to sell, and which sires to use. ## How Genomic Breeding Values Are Calculated Genomic breeding values are derived from a statistical process that links thousands of DNA markers spread across the genome to recorded performance data from a reference population. The reference population consists of animals that have both DNA marker information and high-quality phenotypic records, such as milk yield, fat percentage, [somatic cell](/blog/guides/somatic-cell) count, and fertility traits. The statistical model estimates the effect of each marker on each trait, and these effects are summed across an animal's own marker profile to produce its gEBV. The accuracy of a gEBV depends on the size and relevance of the reference population, the heritability of the trait, and the genetic distance between the tested animal and the reference animals. For production traits with high heritability, such as milk yield, gEBV accuracy can approach that of a progeny test. For low-heritability traits such as fertility or health, accuracy is lower but still useful for ranking animals. The calculation process is performed by national genetic evaluation centers or breed associations. Farmers do not need to compute gEBVs themselves. They receive results from the testing laboratory or the genetic evaluation provider. The key management decision is understanding how to interpret the numbers and how much weight to place on them compared to other information. ## Interpreting Genomic Test Results Genomic test results are typically reported as Predicted Transmitting Abilities (PTAs) or gEBVs, expressed in the units of the trait. For example, a milk yield PTA of +500 kg means the animal is expected to transmit an additional 500 kg of milk per lactation compared to the breed base. A composite index, such as Net Merit or Lifetime Net Merit, combines multiple trait PTAs into a single dollar value that reflects expected lifetime profit. Farmers should examine the reliability or accuracy value that accompanies each gEBV. Reliability ranges from 0 to 1 and indicates how much the prediction is expected to change as more information becomes available. A reliability of 0.40 for a young heifer means the gEBV is moderately reliable but could shift when she has her own lactation records. A reliability of 0.75 or higher is typical for progeny-tested sires. When comparing animals, use the composite index as the primary ranking tool. Then examine individual trait PTAs to identify strengths and weaknesses. For example, a heifer with a high Net Merit index but low fertility PTA may still be a good candidate for breeding if you plan to use sexed semen and manage her carefully. A heifer with average index but excellent health trait PTAs may be valuable for a herd with specific disease challenges. ## Integrating Genomic Data into Herd Selection Decisions The practical workflow for using genomic testing in a dairy herd begins with identifying which animals to test. The highest return on investment typically comes from testing heifer calves that are candidates for replacement. Testing all heifer calves allows you to rank them at birth and cull the lowest-ranking animals before they incur rearing costs. Testing only a subset, such as the top 50 percent based on pedigree, reduces testing costs but may miss valuable animals from lower-ranked parents. Once results are received, create a ranked list using the composite index. Set a minimum index threshold for retention. Animals below the threshold can be sold as beef calves or raised for sale as breeding stock. Animals above the threshold are candidates for breeding. Within the retained group, use individual trait PTAs to match animals to specific breeding strategies. For example, heifers with high fertility PTAs can be bred to high-production sires, while heifers with lower fertility PTAs may benefit from being bred to proven fertility sires. Genomic testing also informs sire selection. When choosing a service sire, examine his gEBVs for the traits that matter most in your herd. If your herd has a high [somatic cell](/blog/guides/somatic-cell) count problem, prioritize sires with low somatic cell score PTAs. If you are expanding, prioritize sires with high daughter pregnancy rate PTAs. The composite index for sires is useful but should be balanced against your herd's specific needs. ## Records and Measurements for Genomic Selection Accurate genomic selection depends on high-quality phenotypic records from the reference population. Farmers contribute to the accuracy of genomic predictions by recording and submitting performance data to the national database. The most important records include: - Milk yield, fat yield, and protein yield from each lactation - Somatic cell count from each test day - Calving ease scores - Gestation length - Daughter pregnancy rate or calving interval - Body condition score - Conformation traits from linear classification Each record should be linked to the animal's unique identification number, such as a breed association registration number or a national animal ID. Without accurate identification, genomic test results cannot be matched to performance records, and the value of testing is reduced. Farmers should also record the date of genomic testing, the laboratory used, and the chip type. This information helps track the reliability of predictions over time and allows comparison with later lactation records. If a heifer's gEBV for milk yield is +300 kg but her first lactation yield is only +50 kg above breed average, the discrepancy may indicate that the reference population did not include animals from similar management systems. ## Common Failure Patterns in Genomic Selection Several common mistakes reduce the effectiveness of genomic testing in dairy herds. The first is testing animals without a clear selection plan. If you test all heifer calves but do not set a culling threshold, you will not capture the economic benefit of early culling. The second is over-relying on genomic predictions for low-heritability traits. For traits like fertility or health, genomic predictions are less accurate than for production traits, and management factors play a larger role. Another failure pattern is ignoring the genetic base. Genomic predictions are expressed relative to a breed base that is updated periodically. A heifer with a PTA of +200 kg for milk yield in 2023 may have a different genetic level than a heifer with the same PTA in 2018. Always check the base year and breed when comparing animals across time or across herds. A fourth failure is testing animals from a population that is genetically distant from the reference population. For example, if you import semen from a breed that has a small reference population in your country, the gEBVs for the resulting calves may have low accuracy. The same issue applies to crossbred animals. Most [genomic prediction](/knowledge/bioinformatics/genomic-prediction-in-livestock-a-decision-framework-for-breeders) equations are developed for purebred populations, and accuracy drops significantly for crossbreds. ## Limitations of Genomic Testing Genomic testing is a powerful tool but has important limitations. The accuracy of gEBVs depends on the size and diversity of the reference population. Small breeds or breeds with limited performance recording will have less accurate predictions. The review "Opportunities and challenges for small populations of dairy cattle in the era of genomics" discusses these limitations for breeds with small populations. Genomic predictions are also less accurate for traits that are influenced by many genes with small effects, such as fertility and health. The study "Genetics and genomics of reproductive performance in dairy and beef cattle" highlights the complexity of reproductive traits and the challenges in predicting them from DNA markers alone. Another limitation is that genomic predictions do not account for genotype-by-environment interactions. An animal that ranks highly in one management system may not rank as highly in a different system. For example, a heifer with high gEBVs for production may not perform well in a low-input grazing system if the reference population was based on high-input confinement herds. Finally, genomic testing cannot predict novel traits that are not recorded in the reference population. If you are selecting for a new trait such as heat tolerance or disease resistance, you need a reference population with records for that trait. The study "Genes and models for estimating genetic parameters for heat tolerance in dairy cattle" describes the genetic basis of heat tolerance and the need for specific data to predict it. ## Welfare and Safety Context Genomic selection can improve animal welfare by enabling selection for health and fertility traits. The study "Symposium review: Why revisit dairy cattle productive lifespan?" discusses the importance of selecting for traits that extend productive lifespan, which reduces involuntary culling and improves welfare. By including health traits in the selection index, farmers can reduce the incidence of mastitis, lameness, and metabolic disorders. However, genomic selection also carries risks if it is focused solely on production traits. Selecting for high milk yield without considering health and fertility can lead to increased metabolic stress, higher disease incidence, and reduced longevity. The study "Systems physiology in dairy cattle: nutritional genomics and beyond" emphasizes the need to consider the whole animal when making genetic decisions. Biosecurity is relevant when collecting DNA samples. Hair roots, blood, and tissue samples should be collected using clean equipment to avoid transmitting pathogens between animals. The study "Spillover of [highly pathogenic avian influenza H5N1](/knowledge/bacteria/avian-bacteria/highly-pathogenic-avian-influenza-h5n1-poultry-surveillance-maps) virus to dairy cattle" reminds us that infectious diseases can emerge in dairy cattle, and any procedure that involves contact with animals should follow biosecurity protocols. Worker safety is also important. When collecting blood samples, use proper restraint and needle disposal procedures. When collecting hair roots, ensure the animal is restrained to avoid injury. Always follow the manufacturer's instructions for sample collection kits. ## Professional Escalation Criteria Farmers should seek professional advice from a geneticist or extension specialist in the following situations: - Genomic test results show unexpected patterns, such as all heifers having very low or very high gEBVs compared to breed average - The reliability of gEBVs is consistently below 0.30 for production traits - You are considering testing crossbred animals or animals from a breed with a small reference population - You want to develop a custom selection index for your herd - You are planning to import semen or embryos from a breed that is not well-represented in your national reference population - You observe a systematic discrepancy between genomic predictions and actual performance across multiple animals A professional can help you interpret results, adjust selection criteria, and decide whether additional testing or alternative breeding strategies are needed. ## Integrating Genomic Data with Herd Health Records: A Practical Decision Framework Genomic testing provides early-life predictions, but its full value is realized only when those predictions are combined with ongoing herd health records to guide management decisions. A structured decision framework that links genomic breeding values to actual health events allows farmers to validate predictions, adjust selection criteria, and improve herd-level outcomes over time. This section presents a practical system for integrating genomic data with health records, including a record-keeping template, a troubleshooting method for discrepancies, and a comparison of selection strategies based on herd health status. ### The Health-Genomic Integration Framework The framework consists of four steps: baseline recording, genomic testing, health event tracking, and decision adjustment. Begin by establishing a baseline of herd health metrics, including somatic cell count, lameness incidence, metabolic disorder rates, and reproductive performance. These metrics provide the context for interpreting genomic predictions. For example, a heifer with a high gEBV for somatic cell score may still be a good candidate if your herd has low mastitis prevalence, but she may be a poor choice if your herd has a chronic mastitis problem. After genomic testing, create a spreadsheet or database that links each animal's gEBVs to her health records. The table below shows a recommended record structure. | Animal ID | Genomic Test Date | Chip Type | Milk Yield gEBV (kg) | Somatic Cell Score gEBV | Daughter Pregnancy Rate gEBV | Composite Index ($) | Health Events (Date and Type) | Management Notes | |-----------|-------------------|-----------|----------------------|--------------------------|------------------------------|---------------------|-------------------------------|------------------| | 1234 | 2024-01-15 | 50K | +450 | +0.15 | +1.2 | +350 | 2024-06-10 Mastitis | Treated, cultured | | 1235 | 2024-01-15 | 50K | +320 | -0.05 | +2.0 | +420 | None | Healthy to date | | 1236 | 2024-01-15 | 50K | +510 | +0.30 | -0.5 | +280 | 2024-07-22 Lameness | Hoof trim needed | This record system allows you to compare predicted values with actual outcomes. If an animal with a low somatic cell score gEBV develops mastitis, investigate whether the infection was environmental or contagious. If an animal with a high daughter pregnancy rate gEBV fails to conceive, check for management factors such as heat detection accuracy or bull fertility. ### Troubleshooting Discrepancies Between Genomic Predictions and Health Outcomes When genomic predictions do not match observed health events, use the following troubleshooting method. First, verify the accuracy of the health record. Was the diagnosis confirmed by a veterinarian? Was the somatic cell count from a test day or a clinical case? Second, check the reliability of the gEBV. A reliability below 0.30 for health traits means the prediction has a wide confidence interval and may shift with more data. Third, consider environmental factors. The study "Genes and models for estimating genetic parameters for heat tolerance in dairy cattle" shows that heat stress can override genetic potential for health traits. If your herd experienced a heat wave during the breeding season, fertility gEBVs may not reflect actual conception rates. Fourth, examine the reference population. If your herd uses a different feeding system or housing type than the reference population, genotype-by-environment interactions may reduce prediction accuracy. The review "[Genomic analysis](/blog/guides/genomic-analysis), progress and future perspectives in dairy cattle selection: A review" notes that genomic predictions are most accurate when the tested animals are managed similarly to the reference population. Fifth, check for parentage errors. A mismatch between an animal's gEBV and its health outcomes may indicate that the DNA sample was mislabeled or that the sire is incorrect. ### Comparison of Selection Strategies Based on Herd Health Status The optimal selection strategy depends on your herd's current health challenges. The table below compares three common scenarios. | Herd Health Status | Primary Selection Goal | Recommended Composite Index Weighting | Additional Trait Focus | Culling Threshold Strategy | |--------------------|------------------------|---------------------------------------|------------------------|----------------------------| | High mastitis prevalence | Reduce somatic cell count | Increase weight on somatic cell score | Udder depth, teat placement | Cull heifers with somatic cell score gEBV above breed average | | Low fertility (calving interval >14 months) | Improve reproductive performance | Increase weight on daughter pregnancy rate | Calving ease, gestation length | Cull heifers with daughter pregnancy rate gEBV below breed average | | Balanced health and production | Maintain current status | Use standard breed index | All traits equally | Cull lowest 20% of composite index | For herds with high mastitis prevalence, prioritize heifers with low somatic cell score gEBVs even if their production gEBVs are average. The study "Extraction of Innate Immune Genes in Dairy Cattle and the Regulation of Their Expression in Early Embryos" suggests that genetic variation in immune function exists and can be selected for. For herds with low fertility, focus on daughter pregnancy rate and calving ease gEBVs. The study "Genetics and genomics of reproductive performance in dairy and beef cattle" confirms that reproductive traits have a genetic component, though heritability is low. ### Records and Measurements for Health-Genomic Integration To implement this framework, maintain the following records for each animal: - Genomic test date, laboratory, and chip density - gEBVs for all health and production traits - Composite index value and rank within the tested group - Health event dates, diagnoses, and treatments - Reproductive records including breeding dates, pregnancy checks, and calving dates - Somatic cell count from each test day - Body condition score at calving and peak lactation - Lameness scores from routine hoof trimming Submit health records to the national database whenever possible. The accuracy of future genomic predictions depends on the size and quality of the reference population. By contributing your herd's health data, you help improve predictions for all farmers. ### Common Failure Patterns in Health-Genomic Integration The most common failure is not recording health events in a format that can be linked to genomic data. If you treat a heifer for mastitis but do not record the date and animal ID, you cannot later compare her health outcome to her somatic cell score gEBV. A second failure is ignoring the reliability of health trait gEBVs. A heifer with a high fertility gEBV but low reliability may still fail to conceive, and you should not assume the prediction is wrong until you have multiple lactations of data. A third failure is using the same selection threshold for all health traits regardless of herd prevalence. If your herd has no lameness problem, you can afford to select heifers with average lameness gEBVs. If lameness is a major issue, you should set a stricter threshold. A fourth failure is not updating the selection index as herd health changes. If you solve your mastitis problem through management changes, you can shift selection emphasis back to production traits. ### Professional Escalation Criteria Seek professional advice from a veterinarian or geneticist if: - You observe a systematic discrepancy between genomic predictions and health outcomes across multiple animals in the same management group - Your herd has a health problem that is not improving despite management changes and genomic selection - You want to develop a custom selection index that weights health traits differently than the standard breed index - You are considering testing for novel health traits such as heat tolerance or disease resistance that are not included in standard genomic panels - You need help setting up a record-keeping system that links genomic data to health records A professional can help you interpret patterns in your data, adjust selection criteria, and decide whether additional testing or alternative breeding strategies are needed. The study "European Dairy Cattle Evaluations and International Use of Genomic Data" describes how different countries handle health trait evaluations and may provide insights for your specific situation. ## Frequently Asked Questions ### What is the difference between a genomic breeding value and a traditional breeding value? A traditional breeding value is estimated from the animal's own performance records and the performance of its relatives. A genomic breeding value is estimated from DNA markers and does not require the animal to have its own performance records. Genomic breeding values can be calculated at birth, while traditional breeding values for traits like milk yield require the animal to complete at least one lactation. ### How accurate are genomic predictions for dairy cattle? Accuracy depends on the trait and the reference population. For high-heritability production traits, accuracy can reach 0.70 to 0.80 for young animals. For low-heritability traits like fertility, accuracy is typically 0.30 to 0.50. Accuracy increases as the reference population grows and as more animals are genotyped. ### How much does genomic testing cost per animal? Costs vary by laboratory, chip density, and volume. Low-density chips are less expensive than high-density chips. Many breed associations and testing companies offer volume discounts. Contact your breed association or a commercial testing laboratory for current pricing. ### Can I use genomic testing on crossbred dairy cattle? Genomic prediction equations are developed for purebred populations. Accuracy for crossbred animals is lower because the marker effects estimated in purebreds may not apply to crossbreds. Some testing companies offer crossbred predictions, but the reliability is reduced. ### How often should I update genomic predictions for my herd? Genomic predictions are updated when the reference population is expanded or when the genetic evaluation model is revised. Most national evaluations are updated annually or semi-annually. You do not need to retest animals unless you want to use a higher-density chip or if the breed association requires a specific chip for registration. ### What traits are included in the composite selection index? The composite index varies by breed and country. Common traits include milk yield, fat yield, protein yield, somatic cell score, daughter pregnancy rate, productive lifespan, calving ease, and conformation traits. Check with your breed association for the specific index used in your region. ### How do I collect a DNA sample for genomic testing? The most common method is to pull hair roots from the tail switch. Ensure the hair follicles are intact. Place the sample in a labeled envelope or tube provided by the testing laboratory. Blood samples can be collected on FTA cards or in EDTA tubes. Tissue samples from ear notches are also used. Follow the laboratory's instructions for sample collection and shipping. ### What should I do if my genomic test results seem wrong? First, check that the animal identification on the test report matches the animal's ear tag or registration number. Second, compare the results to the animal's pedigree. If the results are inconsistent with parent averages, there may be a parentage error. Contact the testing laboratory or your breed association for assistance. ## Related Farming Guides - [Dairy Cow Culling Decisions And Records](/knowledge/animal-farming/dairy-cattle/dairy-cow-culling-decisions-and-records) - [Dairy Cow Cooling System Management](/knowledge/animal-farming/dairy-cattle/dairy-cow-cooling-system-management) - [Beef Cattle Water Quality Testing](/knowledge/animal-farming/beef-cattle/beef-cattle-water-quality-testing) - [How To Design A Comfortable Dairy Cow Barn](/knowledge/animal-farming/dairy-cattle/how-to-design-a-comfortable-dairy-cow-barn) - [Dairy Cattle Farming Nutrition Housing Health Signals And Herd Management](/knowledge/animal-farming/dairy-cattle/dairy-cattle-farming-nutrition-housing-health-signals-and-herd-management) ## Related Clinical & Scientific Guides * [Evaluating Feed Additives for Dairy Cow Performance](/knowledge/animal-farming/dairy-cattle/evaluating-feed-additives-for-dairy-cow-performance) * [Dairy Barn Fire Safety: Design and Prevention Measures](/knowledge/animal-farming/dairy-cattle/dairy-barn-fire-safety-design-prevention) * [Dairy Cow Pregnancy Loss Records and Review](/knowledge/animal-farming/dairy-cattle/dairy-cow-pregnancy-loss-records-and-review) ## References and Further Reading - [www.nrcs.usda.gov](https://www.nrcs.usda.gov/) - [www.merckvetmanual.com](https://www.merckvetmanual.com/management-and-nutrition) - [Genetics and genomics of reproductive performance in dairy and beef cattle.](https://pubmed.ncbi.nlm.nih.gov/24703258). Animal : an international journal of animal bioscience, 2014. - [Spillover of highly pathogenic avian influenza H5N1 virus to dairy cattle.](https://pubmed.ncbi.nlm.nih.gov/39053575). Nature, 2024. - [Systems physiology in dairy cattle: nutritional genomics and beyond.](https://pubmed.ncbi.nlm.nih.gov/25387024). Annual review of animal biosciences, 2013. - [Review: Opportunities and challenges for small populations of dairy cattle in the era of genomics.](https://pubmed.ncbi.nlm.nih.gov/26957010). Animal : an international journal of animal bioscience, 2016. - [Symposium review: Why revisit dairy cattle productive lifespan?](https://pubmed.ncbi.nlm.nih.gov/32089299). Journal of dairy science, 2020. - [European Dairy Cattle Evaluations and International Use of Genomic Data.](https://pubmed.ncbi.nlm.nih.gov/38997844). The Veterinary clinics of North America. Food animal practice, 2024. - [Genetics and genomics of dairy cattle](https://doi.org/10.1016/B978-0-12-817052-6.00006-9). Animal Agriculture Sustainability Challenges and Innovations, 2019. - [Genes and models for estimating genetic parameters for heat tolerance in dairy cattle](https://doi.org/10.3389/fgene.2023.1127175). Frontiers in Genetics, 2023. - [Genomic analysis reveals population structure and selection signatures in plateau dairy cattle](https://doi.org/10.1186/s12864-025-11335-0). [BMC Genomics](/blog/guides/bmc-genomics), 2025. - [Extraction of Innate Immune Genes in Dairy Cattle and the Regulation of Their Expression in Early Embryos](https://doi.org/10.3390/genes15030372). Genes, 2024. - [Genomic analysis, progress and future perspectives in dairy cattle selection: A review](https://doi.org/10.3390/ani11030599). Animals, 2021. > This article is educational and is not a substitute for veterinary diagnosis, treatment, public-health guidance, or regulatory reporting.