- Selection objectives define what a breeding programme is intended to achieve through genetic selection. They provide the biological, economic, and management framework for deciding which traits should be improved, how strongly each trait should be emphasized, and how genetic change should be balanced across production, reproduction, health, welfare, survival, adaptation, and other important characteristics. A clear selection objective is essential because animal breeding rarely aims to maximize a single trait. Instead, successful breeding programmes usually seek to improve several traits simultaneously while controlling undesirable genetic changes, maintaining genetic diversity, and ensuring that animals remain healthy, fertile, functional, and adapted to their production environments.
- The foundation of a selection objective is the distinction between the breeding objective and the information used to select animals. The breeding objective describes the desired direction of genetic change, whereas selection criteria are the measurable sources of information used to identify animals with high genetic merit. For example, a dairy breeding programme may have an objective involving milk yield, milk composition, fertility, longevity, udder health, disease resistance, feed efficiency, and welfare. Some of these objectives may be measured directly, while others may be represented through indicator traits or genetically correlated measurements. The distinction is important because the traits that are easiest to measure are not necessarily the traits that should receive the greatest emphasis in the breeding objective.
- Animal performance can be represented conceptually as P = G + E, where P is phenotypic performance, G is genetic contribution, and E represents environmental effects. Genetic performance itself may include additive genetic effects, dominance effects, and epistatic effects. For selection purposes, additive genetic variation is particularly important because additive genetic effects are transmitted predictably from parents to offspring and therefore determine much of the expected response to selection. A selection objective must therefore be connected to the genetic architecture of the traits being improved and to the extent to which observed differences among animals reflect heritable genetic differences rather than environmental variation.
- A useful selection objective begins with a clear definition of the long-term goals of the breeding programme. These goals may include increasing productivity, improving product quality, increasing fertility, reducing disease incidence, improving survival and longevity, increasing resistance to environmental stress, improving feed efficiency, reducing greenhouse-gas intensity, or improving animal welfare. The relative importance of these goals depends on the species, production system, market, environment, management conditions, and breeding programme. A trait that is highly valuable in one production system may have less importance in another.
- The concept of breeding objective is closely related to the idea of the aggregate breeding value. In quantitative genetics, the breeding objective can be represented as a weighted combination of the breeding values of several traits. A simplified form is:
- H = a₁A₁ + a₂A₂ + … + aₙAₙ
- where H is the aggregate breeding objective, A represents the additive genetic breeding value for each trait, and a represents the relative economic or biological importance assigned to each trait. The weights are not simply arbitrary preferences. In an economic breeding objective, they are commonly based on the expected economic consequences of a unit of genetic change in each trait while accounting for the production system and other relevant biological relationships.
- Economic weights are therefore an important component of many selection objectives. An economic weight describes the marginal value associated with changing a trait by one unit while other relevant traits are held constant. For example, the economic value of increasing milk yield depends not only on additional milk revenue but also on feed costs, labour, health consequences, quota systems, milk composition, and other production factors. Similarly, the economic value of improving fertility may arise from reduced insemination costs, shorter calving intervals, lower replacement costs, increased productive lifetime, and improved reproductive efficiency.
- Economic weighting does not mean that every breeding objective should be determined solely by short-term financial returns. Modern breeding programmes increasingly include non-market traits, such as animal welfare, disease resistance, environmental adaptation, resilience, methane emissions, sustainability, and robustness. These traits may have important biological, ethical, regulatory, environmental, or societal value even when they do not have a simple market price. Consequently, selection objectives can incorporate both economic and non-economic considerations.
- A major principle of selection objectives is that improvement in one trait can affect other traits. Traits may have positive, negative, or near-zero genetic correlations. Genetic correlation can be expressed as:
- r_A = Cov_A(X,Y) / (σ_A,X × σ_A,Y)
- A favourable genetic correlation can allow simultaneous improvement in two traits, whereas an unfavourable correlation can create an important breeding trade-off. For example, selection for increased production may sometimes be genetically associated with changes in fertility, health, body condition, or longevity. The existence of such relationships means that a breeding objective must consider the entire system rather than treating each trait independently.
- Genetic covariance is particularly important when constructing selection objectives because genetic changes in one trait can generate correlated responses in another. A breeding programme that focuses only on the most economically visible trait may unintentionally produce undesirable changes in secondary traits. Including those traits in the objective can help maintain a more balanced direction of genetic improvement.
- The distinction between selection objective and selection criterion is fundamental. The objective describes what the breeding programme wants to improve, while the selection criterion describes the information used to identify animals that are likely to have high merit for that objective. The criterion may include an animal’s own phenotype, relatives’ performance, progeny records, pedigree relationships, genomic information, or a combination of these sources. An animal may therefore be selected using a trait that is not itself part of the final breeding objective if that trait provides useful information about genetic merit for an objective trait.
- This is particularly important when an objective trait is difficult, expensive, late in life, sex-limited, or impossible to measure directly. Indicator traits can provide information about the genetic merit for such traits. For example, genomic information, disease records, physiological measurements, reproductive indicators, or early-life performance may contribute to predicting breeding values for traits that cannot be measured efficiently on every selection candidate.
- The effectiveness of a selection objective depends on the amount of additive genetic variation available in the population. If a trait has genetic variation, selection can potentially change its population mean over generations. However, the magnitude and speed of change depend on heritability, selection intensity, selection accuracy, genetic standard deviation, and generation interval. A simplified relationship for annual genetic improvement is:
- ΔG/year = i × r × σ_A / L
- where i is selection intensity, r is selection accuracy, σ_A is the additive genetic standard deviation, and L is the generation interval. A well-designed selection objective should therefore not only specify what should be improved but should also consider whether sufficient information and genetic variation exist to achieve meaningful improvement.
- Heritability is relevant because it influences how effectively phenotypic information can predict breeding value. Narrow-sense heritability is:
- h² = σ²_A / σ²_P
- where σ²_A is additive genetic variance and σ²_P is phenotypic variance. Traits with higher heritability can often respond efficiently to individual phenotypic selection, whereas traits with low heritability may require information from relatives, repeated records, progeny, genomic data, or multi-trait evaluation. Importantly, low heritability does not mean that a trait should be excluded from the selection objective. A low-heritability trait may have substantial economic, biological, welfare, or sustainability importance and can still respond to selection when appropriate evaluation methods are used.
- Selection objectives frequently include a combination of production traits, growth traits, reproductive traits, fertility traits, health traits, survival and longevity, feed efficiency, behavioural traits, welfare-related traits, and adaptation traits. In livestock, production traits may include milk yield, meat production, carcass composition, egg production, wool production, fibre quality, or growth rate. Reproductive objectives may include age at sexual maturity, conception rate, litter size, calving interval, semen quality, or survival of offspring. Health objectives may include disease resistance, immune function, udder health, hoof health, metabolic health, and resistance to specific diseases.
- Functional traits can be particularly important because improvements in production do not automatically guarantee improvements in biological efficiency or animal performance over the entire lifetime. Fertility, survival, longevity, health, temperament, structural soundness, and disease resistance can influence the sustainability and profitability of production systems. Including these traits in the selection objective can help prevent excessive emphasis on short-term production at the expense of overall animal function.
- Animal welfare can also be incorporated into selection objectives. Genetic selection may influence traits associated with health, structural soundness, behaviour, temperament, disease resistance, reproductive ability, and resilience. However, welfare should not be treated as purely genetic. Housing, nutrition, handling, stocking density, veterinary care, environmental conditions, and management practices remain essential determinants of animal welfare. Genetic selection should therefore complement, rather than replace, good management.
- Adaptation and resilience are increasingly important selection objectives. Animals may face heat stress, cold, disease challenge, variable feed availability, drought, humidity, parasites, or other environmental pressures. Selection for adaptation can improve performance under challenging conditions, while selection for resilience aims to improve an animal’s ability to maintain relatively stable performance and recover from environmental disturbances. These objectives can become particularly important as production environments change.
- The selection objective should also consider genotype–environment interaction (G×E). Genetic rankings can differ across environments when animals respond differently to environmental conditions. A genotype that performs exceptionally well in a high-input production system may not be optimal in a low-input, hot, disease-challenged, or pasture-based system. Therefore, selection objectives should be aligned with the environments in which animals and their descendants are expected to perform.
- For breeding programmes operating across several environments, multi-environment evaluation can help identify animals with broad adaptation or environment-specific merit. In some situations, the breeding objective may explicitly assign value to stability, robustness, or performance under particular environmental conditions rather than simply maximizing performance in an average environment.
- One of the most important applications of selection objectives is the development of a selection index. A selection index combines information from several traits or information sources into a single numerical value that can be used to rank selection candidates. A simplified index can be written as:
- I = b₁x₁ + b₂x₂ + … + bₙxₙ
- where I is the selection index, x represents measured information, and b represents index coefficients. The coefficients are determined from the relationships among information sources, genetic parameters, and the breeding objective. A well-designed index therefore allows multiple traits to be considered simultaneously while accounting for differences in heritability, genetic covariance, economic importance, and information accuracy.
- Selection index theory is particularly useful when traits have different units and scales. Milk yield may be measured in kilograms, fertility may be measured as a percentage or probability, disease resistance may involve binary records, and longevity may be measured in days or years. A selection index provides a systematic method for combining these different types of information into a coherent ranking.
- Modern genetic evaluation systems frequently use BLUP, animal models, and genomic information to estimate breeding values for traits included in or related to the selection objective. Traditional pedigree-based evaluations use the numerator relationship matrix, commonly represented as the A matrix, to model expected genetic relationships among animals. Genomic evaluation can incorporate a genomic relationship matrix, often represented as the G matrix, allowing breeding values to be predicted using genomic information.
- Genomic selection can increase selection accuracy, especially for young animals that have little or no own performance or progeny information. Genomic selection can therefore allow breeding programmes to place greater emphasis on difficult-to-measure traits, health traits, reproductive traits, and other traits for which traditional progeny testing may be slow or expensive. However, genomic prediction depends on the quality and relevance of the reference population, phenotypic data, genotypes, pedigree information, and validation procedures.
- Progeny testing remains valuable when traits are difficult to measure directly on selection candidates or when offspring performance provides substantial information about parental breeding value. Family information, within-family comparisons, repeated records, and combined selection can also contribute to the evaluation of candidates. Modern programmes often integrate these sources rather than relying on a single selection method.
- Selection objectives can include threshold traits, such as disease status, pregnancy success, survival, calving difficulty, or other categorical outcomes. Such traits are often analysed using threshold or generalized statistical models because the observed phenotype represents categories generated by an underlying continuous liability. This allows genetic evaluation to incorporate biologically meaningful traits even when the recorded phenotype is binary or categorical.
- The design of a selection objective must also account for genetic antagonisms. An antagonistic relationship occurs when improvement in one trait tends to cause an undesirable change in another. Such relationships are not necessarily reasons to abandon selection for the first trait. Instead, they indicate that the breeding objective should explicitly include both traits so that the overall direction of genetic change can be optimized.
- For example, if selection for increased production is associated with reduced fertility, a breeding programme may include both production and fertility in the objective. The resulting selection strategy may produce slightly less production gain than selection for production alone but may generate greater overall biological and economic value by preventing deterioration in fertility.
- The same principle applies to health and disease resistance. If selection strongly increases production but simultaneously increases susceptibility to disease or metabolic disorders, the long-term value of the production improvement may be reduced. Including health traits in the breeding objective can encourage genetic improvement that is more sustainable and compatible with animal health.
- Multiple-trait selection is therefore closely connected to selection objectives. The purpose is not necessarily to maximize every trait independently. Instead, the objective is to achieve an appropriate combination of genetic changes that produces the greatest overall value under the breeding programme’s goals and constraints. This may require compromises among traits, particularly when genetic correlations are unfavourable.
- Selection objectives also influence the management of genetic diversity. Intensive selection can increase genetic concentration when a small number of highly ranked animals contribute disproportionately to the next generation. Excessive use of popular sires can increase relatedness and accelerate the accumulation of inbreeding. A selection objective should therefore be evaluated not only by expected short-term genetic gain but also by its consequences for long-term population sustainability.
- The expected inbreeding coefficient of offspring from a mating can be related to parental coancestry:
- E(F_offspring) = φ(sire, dam)
- where φ represents the coancestry or kinship coefficient between the parents. For close relatives, the expected offspring inbreeding can also be expressed in relation to their coefficient of relationship. In general, relationship is approximately twice kinship:
- r ≈ 2φ
- These relationships are important because a breeding programme can achieve high genetic gain while simultaneously increasing the risk of genetic concentration if mating decisions are not controlled.
- Effective population size (Ne) provides another measure of the rate at which genetic diversity may be lost. A simplified approximation for the increase in inbreeding per generation is:
- ΔF ≈ 1 / (2Ne)
- Maintaining an adequate effective population size can therefore be an important long-term objective of sustainable breeding programmes. Genetic diversity allows populations to retain adaptive potential and provides genetic variation for future selection.
- Optimal contribution selection (OCS) provides one approach for balancing genetic gain and genetic diversity. Instead of selecting animals solely according to their estimated breeding values, OCS determines appropriate genetic contributions from selected animals while controlling coancestry and the expected rate of inbreeding. This allows a breeding programme to pursue genetic improvement while avoiding excessive concentration of ancestry.
- Mate allocation can further support the selection objective after breeding candidates have been selected. Rather than simply selecting the best animals and mating them randomly, mating plans can consider genetic merit, relatedness, carrier status for deleterious variants, genomic relationships, and other constraints. The goal is to create offspring with high expected genetic merit while managing inbreeding and genetic risk.
- Genomic relatedness and runs of homozygosity (ROH) can provide additional information for monitoring genetic diversity. The proportion of the autosomal genome contained in runs of homozygosity can be summarized as:
- F_ROH = Total length of ROH / Total autosomal genome length
- This measure can provide information about recent and historical autozygosity and can complement pedigree-based estimates of inbreeding.
- Selection objectives can also incorporate information about deleterious genetic variants. Genetic testing can identify animals carrying known recessive variants, allowing breeders to design mating strategies that reduce the probability of producing affected offspring without unnecessarily removing valuable genetic diversity from the population. Complete elimination of every potentially harmful variant may not always be appropriate because genetic variants can have different effects, frequencies, penetrance, and relationships with other traits.
- The development of a selection objective should therefore involve careful definition of the breeding population, production system, time horizon, market environment, management conditions, and biological constraints. The objective should be sufficiently stable to support long-term genetic improvement but flexible enough to adapt when economic conditions, environmental pressures, regulations, technologies, or societal priorities change.
- A short-term selection objective may emphasize immediate production or economic performance, whereas a long-term breeding objective should consider the cumulative consequences of genetic change over many generations. Traits such as longevity, disease resistance, fertility, adaptation, and genetic diversity may produce benefits that become increasingly important over time even if their immediate economic value is less obvious.
- The time horizon is especially important because genetic selection is cumulative and largely irreversible in the short term. Once a population has been genetically shifted toward a particular combination of traits, reversing undesirable changes may require many generations. Consequently, selection objectives should be designed with both current performance and future population health in mind.
- Data quality is another critical component. Selection objectives are only useful when the genetic evaluation system is supported by reliable phenotypic records, accurate pedigree information, appropriate contemporary groups, consistent trait definitions, and sufficient genetic connectedness among herds, flocks, farms, or populations. Poor-quality records can reduce selection accuracy and lead to misleading estimates of breeding value.
- Contemporary groups are particularly important because animals should generally be compared under similar environmental and management conditions when evaluating phenotypic performance. Differences caused by feed, housing, disease exposure, season, age, management, or other environmental factors should not be incorrectly interpreted as genetic differences.
- Repeated measurements can also improve evaluation for traits that vary over time. Repeatability describes the extent to which repeated records on the same animal are correlated because of permanent effects. Multiple records can improve the estimation of an animal’s long-term performance and breeding value when appropriately modelled.
- Maternal effects, common environmental effects, and permanent environmental effects may also influence observed performance. For example, offspring raised by the same dam may share maternal genetic and environmental influences. If these effects are ignored, breeding values can be biased and the resulting selection objective may not be achieved as intended.
- Selection objectives should also be monitored through the actual realized genetic response of the population. A breeding programme should compare observed genetic trends with the expected response and investigate deviations. If a trait is not improving as expected, possible explanations include insufficient genetic variation, low selection accuracy, inadequate recording, environmental changes, unfavourable genetic correlations, low selection intensity, or inappropriate weighting of the trait.
- The selection objective can therefore be viewed as a dynamic management tool rather than simply a list of traits. It connects biological goals, economic priorities, genetic parameters, statistical evaluation, selection decisions, mating strategies, and long-term population management. A strong objective provides a clear direction for the entire breeding programme.
- An effective selection objective should be balanced, measurable, genetically achievable, economically or biologically justified, and sustainable over generations. It should avoid excessive emphasis on a single trait when that emphasis creates undesirable correlated responses. It should also recognize that breeding programmes operate within real production environments where health, fertility, survival, adaptation, welfare, and management are interconnected.
- Modern animal breeding increasingly moves toward balanced breeding objectives that combine production efficiency with health, fertility, longevity, welfare, environmental adaptation, resilience, and genetic diversity. Genomic selection, advanced statistical models, automated phenotyping, precision livestock technologies, and improved recording systems are expanding the range of traits that can be incorporated into breeding programmes.
- The ultimate purpose of a selection objective is therefore not simply to produce animals with the highest value for one measurable trait. It is to guide genetic change toward a population that performs efficiently, remains healthy and fertile, adapts to its environment, supports animal welfare, retains sufficient genetic diversity, and remains valuable and sustainable over many generations. When properly designed, selection objectives provide the foundation for selection indexes, multiple-trait selection, BLUP, genomic selection, mate allocation, optimal contribution selection, and other modern breeding strategies. They transform broad breeding goals into a structured framework for achieving predictable and sustainable genetic improvement.
