- Individual selection is a method of artificial selection in which animals are chosen for breeding primarily according to their own phenotypic performance for one or more traits. The basic principle is simple: animals that perform better for economically, biologically, or biologically relevant breeding objectives are given a greater opportunity to reproduce. Individual selection is one of the oldest and most straightforward methods of animal breeding and remains important in livestock, poultry, aquaculture, companion animal, laboratory animal, and conservation breeding programs. Although modern breeding programs increasingly use BLUP, pedigree information, genomic selection, and other sources of information, individual performance remains an important component of genetic evaluation because the animal’s own record can contain valuable information about its genetic potential.
- The fundamental genetic principle behind individual selection is that an observed phenotype is influenced by both genetic and environmental factors. This relationship can be expressed as P = G + E, where P is the observed phenotype, G is the genetic component, and E represents environmental influences. The genetic component may include additive genetic effects, dominance effects, and epistatic effects. For selection purposes, additive genetic effects are especially important because they contribute to the animal’s breeding value, which represents the expected genetic contribution of an individual to its offspring. An animal may have an excellent phenotype because of superior genes, favorable management, exceptional nutrition, temporary environmental conditions, or a combination of these factors. Individual selection is therefore most effective when phenotypic differences accurately reflect genetic differences.
- The relationship between phenotype and breeding value is strongly influenced by heritability. Narrow-sense heritability is defined as h² = σ²_A / σ²_P, where σ²_A is additive genetic variance and σ²_P is phenotypic variance. When heritability is relatively high, differences among animals are more strongly associated with differences in additive genetic merit, making individual selection more effective. When heritability is low, environmental effects account for a larger proportion of phenotypic variation, and selecting animals solely on their own performance can be less accurate. Low heritability, however, does not mean that genetic improvement is impossible. It means that additional information, such as family records, progeny records, repeated records, pedigree relationships, or genomic information, may be especially valuable.
- The expected response to selection can be expressed in its simplest form as R = h² × S, where R is the response to selection and S is the selection differential. The selection differential is the difference between the mean phenotype of selected animals and the population mean. If the selected animals have substantially better phenotypic performance than the population average and the trait has sufficient heritable variation, the next generation is expected to improve genetically. The actual response in a breeding population can differ from this simple expectation because of selection accuracy, environmental changes, genetic correlations, population structure, genotype–environment interaction, and changes in the breeding population over time.
- Individual selection can be particularly effective for traits that are highly heritable, can be measured accurately on the candidate animal, and can be recorded before the animal reaches reproductive age. Examples may include certain body measurements, growth traits, wool characteristics, carcass-related traits measured directly or indirectly, some production traits, and other traits for which the animal’s own performance provides a useful indication of its breeding value. The method is less straightforward for traits that are expressed only in one sex, appear late in life, require destructive measurement, have low heritability, or are strongly influenced by environmental conditions.
- The quality of the individual record is therefore extremely important. Animals should ideally be evaluated under comparable management conditions, with accurate measurements and appropriate adjustment for age, sex, contemporary group, litter or hatch effects, season, location, nutritional conditions, and other systematic environmental influences. For example, comparing the body weights of animals raised under substantially different nutritional conditions may lead to incorrect conclusions about genetic merit. Standardized recording and appropriate statistical adjustment help separate genetic differences from environmental differences.
- Contemporary groups are especially important in practical individual selection. Animals that are raised together or under similar environmental conditions can be compared more fairly because they experience many of the same environmental influences. Modern genetic evaluation systems go further by accounting statistically for multiple environmental and genetic factors simultaneously. This reduces the risk that an environmental advantage is incorrectly interpreted as superior genetic merit.
- Individual selection is closely related to the concept of selection differential. If the population mean for a trait is 100 units and the selected animals average 110 units, the selection differential is 10 units. However, the phenotypic difference does not necessarily translate directly into the same genetic change. The expected response depends on heritability and the accuracy with which genetic merit is identified. Consequently, selection based on phenotype should not be interpreted as selecting genes directly; it is selecting animals using phenotypic evidence about their genetic merit.
- Selection intensity also affects genetic change. If only a small proportion of animals are selected for reproduction, the difference between selected animals and the population mean may be large, increasing selection intensity. If many animals are retained, selection intensity is lower. Selection intensity must be balanced against reproductive requirements, population size, welfare, genetic diversity, and long-term sustainability. Extremely intense selection may produce rapid short-term genetic progress but can also increase unequal reproductive contribution, genetic concentration, and potentially the rate of inbreeding.
- The generation interval is another important consideration. If animals can be evaluated accurately and selected at a younger age, genetic improvement can potentially occur more rapidly per unit of time. The rate of genetic improvement can be represented approximately as 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 generation interval. Individual selection can contribute to shorter generation intervals when traits can be measured early. However, reducing generation interval should not be pursued at the expense of health, fertility, longevity, welfare, or genetic diversity.
- Individual selection can be based on a single trait or incorporated into a multiple-trait selection program. In real breeding programs, animals are rarely evaluated for only one characteristic. A breeding objective may include growth, production, fertility, disease resistance, survival, feed efficiency, behaviour, welfare, adaptation, and other traits. These traits may have unfavorable or favorable genetic correlations with one another. Selecting strongly for one trait can therefore unintentionally change another trait. For example, intense selection for production without considering fertility or health can create undesirable correlated responses when the traits have antagonistic genetic relationships.
- A selection index can be used to combine information from several traits according to their economic or biological importance. Rather than selecting animals simply because they rank highest for one phenotype, an index can combine estimated genetic merit across multiple traits. This makes individual selection more compatible with balanced breeding objectives. The approach is particularly useful when breeding programs need simultaneous improvement in production, reproduction, health, welfare, and adaptation.
- The distinction between phenotypic selection and individual selection is important. Phenotypic selection is a broad concept describing selection based on observable traits, whereas individual selection specifically emphasizes the candidate animal’s own performance. Family selection instead uses information from relatives, while progeny testing uses information from offspring. Modern breeding programs can combine all these sources rather than treating them as completely separate methods.
- One major advantage of individual selection is its simplicity. The breeder can directly observe or measure the animal and make a selection decision without waiting for offspring performance. This can reduce management complexity and, for suitable traits, reduce the generation interval. Individual selection can also be relatively inexpensive when the trait is easy to measure. In small breeding populations or systems with limited access to advanced genetic evaluation infrastructure, individual selection may remain a practical method.
- However, individual selection has important limitations. The most obvious is that the phenotype is not the same as breeding value. Environmental variation can cause an animal with average genetic merit to appear superior, while an animal with excellent genetic potential may appear average under unfavorable conditions. This problem becomes greater when traits have low heritability. The method can also be ineffective when the trait cannot be measured directly on selection candidates or when the trait is expressed only after reproductive age.
- Sex-limited traits provide another important example. Milk production in dairy females is expressed primarily in females, so a young male cannot be selected directly according to his own milk yield. Similarly, egg production cannot be directly evaluated from the male’s own phenotype. In such cases, relatives, progeny, pedigree information, or genomic information become particularly important. Individual selection may still contribute through correlated traits or other measurements, but selection based solely on the candidate’s own phenotype is insufficient.
- Traits that are difficult or expensive to measure also create limitations. Feed efficiency, methane emissions, disease resistance, fertility, longevity, and some welfare traits may require specialized facilities, long observation periods, repeated measurements, or large amounts of data. Individual selection based on a single record may therefore be less reliable than a combined evaluation incorporating family, progeny, pedigree, and genomic information.
- Repeated records can improve the usefulness of individual information for traits that can be measured multiple times. For example, production or behavioural measurements may be recorded repeatedly on the same animal. Repeated records can help distinguish permanent differences among animals from temporary environmental fluctuations. This is related to repeatability, which describes the consistency of repeated measurements of the same trait in an individual.
- Maternal effects can also complicate individual selection. In young animals, early growth may be influenced not only by the animal’s own genes but also by maternal genotype, milk production, uterine environment, maternal behaviour, litter environment, and other maternal factors. An animal that grows rapidly because it received an exceptionally favorable maternal environment may not necessarily transmit that advantage genetically to its own offspring. Statistical models that account for maternal effects can therefore improve selection decisions.
- Similarly, common environmental effects can make relatives resemble one another even when the similarity is not entirely genetic. Full-sibs, for example, may share the same dam, early environment, management, nutrition, and disease exposure. If their shared environmental effects are not considered, the apparent superiority of an individual or family may be incorrectly attributed to genetics.
- Modern animal breeding increasingly combines individual performance with other sources of information. BLUP and animal models can incorporate an animal’s own records, relatives’ records, pedigree relationships, and environmental effects to estimate estimated breeding values (EBVs). This provides a more informative basis for selection than simply ranking animals by raw phenotype.
- The pedigree relationship matrix, commonly represented by the A matrix, describes expected genetic relationships based on pedigree information. Genomic information can provide additional information through the G matrix, which reflects realized genomic similarity based on genetic markers. Combining individual phenotypic records with pedigree and genomic information can substantially improve the accuracy of genetic evaluation, particularly when the candidate has limited own performance or limited progeny information.
- Genomic selection has transformed the role of individual selection. With dense genetic marker information, animals can receive genomic estimated breeding values (GEBVs) before they have extensive phenotypic records or progeny. This allows breeders to select young animals using information about their genome together with phenotypic and pedigree information. The resulting approach is not a replacement for individual performance records; rather, genomic selection works best when supported by high-quality phenotypic data from appropriate reference populations.
- Family information can be particularly useful when the individual phenotype has limited accuracy. The performance of full-sibs, half-sibs, parents, and other relatives provides additional evidence about an individual’s genetic merit. This is why family selection, within-family selection, and combined selection are important complements to individual selection.
- Within-family selection can be especially useful when family members share environmental conditions. Comparing individuals within the same family can reduce some environmental differences between families and focus more strongly on differences among relatives. However, family structure, maternal effects, common environmental effects, and family size must be considered carefully.
- Progeny testing provides another source of information. When a trait is difficult to measure directly in the candidate or is expressed in only one sex, offspring performance can provide evidence about the candidate’s breeding value. The disadvantage is that progeny testing can require substantial time and resources and can increase the generation interval. Modern genomic selection can reduce some of this delay by providing genetic information at an earlier age.
- Individual selection also interacts with genetic diversity. If the same small group of high-performing animals is repeatedly selected and used extensively, genetic contributions can become concentrated. This may increase the popular sire effect, reduce effective population size, and increase the rate of inbreeding. The approximate relationship ΔF ≈ 1 / (2Ne) illustrates why maintaining an adequate effective population size (Ne) is important. The exact relationship in real breeding populations can be more complex because reproductive contribution is often unequal and populations may not behave as idealized random-mating populations.
- A breeding program should therefore distinguish between selecting the best animals and allowing those animals to make unlimited genetic contributions. Selection determines which animals have higher reproductive opportunity, whereas mating management determines how their genes are combined and how much each animal contributes to future generations. Tools such as mate allocation, mean kinship, and optimal contribution selection can help maintain genetic diversity while retaining desirable genetic merit.
- The expected inbreeding of an offspring is related to the genetic relationship between its parents. If φ(sire, dam) represents their coefficient of coancestry, then E(F_offspring) = φ(sire, dam). When relationship is defined as twice the coancestry, the relationship can be expressed as E(F_offspring) = r(sire, dam) / 2. These relationships can be incorporated into mating decisions so that high genetic merit is achieved without unnecessarily mating closely related animals.
- Genomic relatedness can further improve management of genetic diversity because pedigree relationships represent expected relatedness, whereas genomic information can reveal variation in the actual DNA inherited by individuals. This can be particularly valuable in populations with incomplete pedigrees, uncertain parentage, or complex ancestry. Genomic information can also identify runs of homozygosity (ROH), which provide information about segments of the genome that are identical by descent.
- The proportion of the autosomal genome contained in ROH can be summarized as F_ROH = Total length of ROH / Total autosomal genome length. ROH patterns can provide additional evidence about recent or historical autozygosity and can complement pedigree-based measures of inbreeding. This information can be considered when selecting animals and designing mating plans.
- Individual selection can also influence genetic concentration. When selection repeatedly favors a small number of superior animals, their descendants may represent an increasing proportion of the population. This can accelerate genetic progress for the selected trait but may reduce the number of independent genetic lineages contributing to future generations. Breeding programs therefore need to monitor both genetic merit and genetic contribution.
- The issue becomes particularly important in populations using artificial insemination, embryo transfer, or other reproductive technologies. These technologies can increase the number of offspring produced by genetically superior animals and therefore increase the speed of genetic dissemination. At the same time, excessive use of a small number of elite animals can increase relatedness and reduce effective population size. Genetic improvement and reproductive management should therefore be considered together.
- Genetic testing can also complement individual selection. Some animals with desirable phenotypes may carry deleterious recessive variants that are not expressed in heterozygous form. If such animals are used extensively, harmful alleles can spread through the population. DNA testing can identify known variants and support mating decisions that avoid producing affected offspring while retaining useful genetic diversity.
- Individual selection should also consider genotype–environment interaction (G×E). An animal that performs exceptionally well in one production environment may not have the same advantage under another climate, feeding system, disease challenge, housing system, or management regime. Selection programs intended for diverse production environments should therefore consider whether genetic rankings remain stable across environments.
- This is particularly important for adaptation traits, including heat tolerance, disease resistance, resilience, feed efficiency under limited resources, and survival under challenging conditions. A breeding program that selects only for performance under highly controlled conditions may unintentionally reduce adaptation to commercial or changing environments. Balanced selection should therefore consider both production and fitness-related traits.
- Individual selection can be applied to health and disease-related traits, but careful interpretation is required. Disease resistance, disease susceptibility, disease tolerance, immune function, survival, and resilience are not identical traits. A single observed disease outcome may reflect pathogen exposure, vaccination, management, nutrition, environmental conditions, and genetic differences. Appropriate phenotyping and statistical models are therefore necessary before using such records for selection.
- For binary or categorical traits, such as whether an animal developed a disease or became pregnant, the underlying genetic liability may be continuous even though the observed outcome is not. Threshold models are often useful for analyzing such traits. Individual selection can still use these observations, but direct ranking of animals from a small number of binary outcomes may be less informative than a genetic evaluation that incorporates relatives, repeated records, and environmental factors.
- Behavioural and welfare-related traits can also be included in individual selection when reliable measurements are available. Temperament, handling response, aggression, fearfulness, social behaviour, and other traits may have genetic components, but they are often strongly affected by management and experience. Selection should therefore use standardized measurement protocols and avoid interpreting a single behavioural observation as a complete representation of genetic merit.
- One of the most important advantages of modern breeding programs is that individual selection no longer needs to operate in isolation. Combined selection can integrate an animal’s own performance with family records, progeny information, pedigree relationships, genomic information, and other relevant data. This approach can improve selection accuracy and reduce some of the limitations of relying exclusively on phenotype.
- The relative value of individual selection therefore depends on the trait, population, recording system, heritability, age of measurement, environmental influence, selection objective, and available genetic information. It is most powerful when the trait is accurately measured on the candidate, expressed early enough to influence selection decisions, and sufficiently heritable. It becomes less reliable when environmental effects dominate, the trait is expressed late, or the candidate’s own phenotype provides little information about its breeding value.
- A sustainable breeding strategy should not simply maximize the performance of the highest-ranking individuals. It should consider genetic gain, fertility, health, survival, longevity, welfare, adaptation, genetic diversity, and long-term population structure. Selection decisions should therefore be evaluated over multiple generations rather than only by short-term performance.
- Individual selection remains a fundamental component of animal breeding because it provides direct information about the candidate animal and can be simple, rapid, and effective for appropriate traits. Its limitations are equally important: phenotype is not identical to breeding value, environmental effects can distort rankings, and individual records may be insufficient for complex, low-heritability, sex-limited, late-expressed, or expensive-to-measure traits. Modern breeding programs address these limitations by combining individual records with family selection, progeny testing, BLUP, pedigree information, genomic selection, and multi-trait genetic evaluation.
- Ultimately, the most effective use of individual selection is not to treat it as an isolated breeding method but as one component of an integrated genetic improvement system. When combined with accurate phenotyping, appropriate statistical evaluation, genomic information, balanced breeding objectives, and responsible management of genetic diversity, individual selection can contribute substantially to sustainable genetic improvement. The long-term goal is not simply to select animals that look superior today, but to identify animals with desirable heritable genetic merit and use them in a way that improves the population while maintaining health, fertility, welfare, adaptation, and genetic diversity for future generations.
