- Genetic correlations between traits describe the degree to which genetic factors affecting one trait are associated with genetic factors affecting another trait. In animal breeding, traits rarely behave as completely independent characteristics. The same genes, or groups of genes, may influence several traits simultaneously, creating a genetic association between them. Understanding genetic correlation is therefore essential for predicting correlated response to selection, designing multiple-trait selection programs, developing selection indexes, and achieving balanced genetic improvement.
- A genetic correlation specifically describes the relationship between the additive genetic values of two traits. It is different from a simple phenotypic correlation because phenotypic observations are influenced by both genetic and environmental factors. Genetic correlation is important for breeding because additive genetic effects are transmitted from parents to offspring and therefore determine much of the predictable response to selection.
- The phenotype of an animal can be represented as P = G + E, where P is phenotypic value, G is genetic value, and E represents environmental effects. Genetic value can contain additive genetic effects, dominance effects, and epistatic effects, but genetic correlation used in quantitative breeding programs generally focuses on the association between additive genetic effects of two traits. This additive relationship determines how selection on one trait can change the genetic level of another trait.
- For two traits X and Y, the additive genetic correlation can be expressed as r_A = Cov_A(X,Y) / (σ_A,X × σ_A,Y), where Cov_A(X,Y) is the additive genetic covariance between the traits, and σ_A,X and σ_A,Y are the additive genetic standard deviations of the two traits. The value of genetic correlation normally ranges from -1 to +1. A value close to +1 indicates a strong positive genetic association, a value close to -1 indicates a strong negative genetic association, and a value near zero indicates little linear additive genetic association.
- A positive genetic correlation means that genes associated with higher genetic values for one trait tend, on average, to be associated with higher genetic values for the other trait. For example, if growth rate and body weight have a positive genetic correlation, selection for increased growth rate may also increase body weight. A positive correlation can therefore produce a favorable correlated response when both traits are desirable, although it can also create an undesirable increase in a trait if excessive values are not preferred.
- A negative genetic correlation means that genes associated with increasing one trait tend to be associated with decreasing the other. This type of relationship can create an antagonistic genetic relationship. For example, selection for increased production may sometimes be genetically associated with reduced fertility or longevity. When such relationships occur, selection for only one trait can lead to an unfavorable correlated response in another trait.
- A genetic correlation close to zero indicates that the additive genetic factors affecting the two traits have little linear association. In such a situation, selection for one trait is expected to produce relatively little correlated genetic change in the other trait, assuming the relevant genetic parameters are accurately estimated. However, a genetic correlation of zero does not necessarily mean that the traits have no biological relationship; it only indicates little linear additive-genetic association.
- Genetic covariance provides the underlying covariance component from which genetic correlation is derived. Genetic covariance can be positive, negative, or close to zero. Genetic correlation standardizes this covariance by the genetic standard deviations of the two traits, allowing genetic relationships to be compared across traits measured on different scales.
- Genetic correlations are fundamentally different from phenotypic correlations and environmental correlations. Phenotypic correlation describes the association between observed phenotypes, while environmental correlation describes the association between environmental effects affecting different traits. Two traits may have a strong phenotypic correlation but a weak genetic correlation if their observed association is largely caused by common environmental or management factors. Conversely, a relatively weak phenotypic association can coexist with a meaningful genetic correlation.
- This distinction is extremely important in animal breeding. If animals receiving better nutrition have both higher body weight and greater fertility, the two traits may appear positively correlated phenotypically even if their additive genetic relationship is weak. Genetic evaluation methods attempt to separate genetic effects from environmental effects so that breeders can make selection decisions based on inherited genetic merit rather than environmental associations.
- The magnitude of genetic correlation has important consequences for selection response. When two traits have a strong positive genetic correlation, selection for one trait can result in substantial positive correlated response in the other. When the genetic correlation is strongly negative, selection for one trait can result in an unfavorable response in the other. The greater the absolute value of the genetic correlation, the greater the potential for correlated genetic change, provided that sufficient additive genetic variation and selection accuracy exist.
- The concept of correlated response is especially important when a trait is difficult, expensive, late-expressed, sex-limited, or impossible to measure directly in all animals. An easily measured indicator trait can sometimes be used for selection if it has a sufficiently strong genetic correlation with the target trait. For example, a measurable indicator trait may provide information about feed efficiency, disease resistance, carcass quality, or fertility when direct measurement of the target trait is difficult.
- The relationship between genetic correlation and correlated response can be illustrated using the simplified expression CR_Y = i × r_X × r_A × σ_A,Y, where CR_Y represents the correlated response in trait Y, i is selection intensity, r_X is the accuracy of selection for trait X, r_A is the genetic correlation between X and Y, and σ_A,Y is the additive genetic standard deviation of trait Y. This expression demonstrates why genetic correlation is a central parameter in predicting indirect genetic change.
- Heritability also affects the efficiency of selection. Heritability is expressed as h² = σ²_A / σ²_P, where σ²_A is additive genetic variance and σ²_P is phenotypic variance. Traits with higher heritability generally provide more information about an individual’s genetic merit from its own phenotype. For low-heritability traits, information from relatives, progeny, repeated records, correlated traits, pedigree relationships, and genomic information can substantially improve genetic evaluation.
- Genetic correlations are estimated using information from pedigrees, phenotypic records, relatives, progeny, and increasingly genomic data. Large and well-structured datasets are important because estimates of genetic correlations can be uncertain when sample sizes are small or when the traits are measured on different subsets of animals. Accurate estimation also requires appropriate statistical models that account for environmental effects, contemporary groups, fixed effects, and relationships among animals.
- In modern genetic evaluation, BLUP and multi-trait animal models are commonly used to estimate breeding values and genetic parameters. A simplified animal model can be written as y = Xb + Za + e, where y represents observations, b represents fixed effects, a represents random additive genetic effects, and e represents residual effects. In a multi-trait model, genetic variances and covariances among traits are estimated simultaneously, allowing genetic correlations to be incorporated into breeding-value prediction.
- Multi-trait genetic evaluation can be particularly valuable when one trait has limited records but is genetically correlated with another trait that has abundant records. Information from the better-recorded trait can contribute to the prediction of genetic merit for the less-recorded trait. This principle is important for traits such as fertility, disease resistance, longevity, and other characteristics that may be difficult or expensive to measure.
- Genomic selection has further expanded the use of genetic correlations. Genomic information can improve the prediction of breeding values, especially when combined with phenotypic and pedigree information in a suitable reference population. Multi-trait genomic evaluation can use genetic relationships between traits to improve prediction for traits with limited phenotypic information. Genomic estimated breeding values (GEBVs) can therefore contribute to more accurate selection decisions for several traits simultaneously.
- Genetic correlations are particularly important when developing breeding objectives. A breeding objective may include production, reproduction, health, survival, welfare, feed efficiency, and adaptation traits. These traits can have favorable or unfavorable genetic relationships. Ignoring those relationships can result in unexpected genetic changes and can make it difficult to achieve the desired overall breeding goal.
- A selection index provides a practical framework for managing genetic correlations among multiple traits. It can be expressed as I = b₁x₁ + b₂x₂ + … + bₙxₙ, where x represents information used for selection and b represents index weights. By incorporating genetic and economic relationships among traits, a selection index can balance improvement across several characteristics instead of maximizing a single trait.
- The breeding objective can be represented as H = a₁A₁ + a₂A₂ + … + aₙAₙ, where A represents additive genetic merit for each trait and a represents the corresponding economic or strategic weight. Genetic correlations influence the relationship between the information used for selection and the underlying breeding objective. Correctly accounting for these correlations is therefore essential for predicting the expected response to selection.
- In livestock breeding, important genetic relationships can occur among growth traits, body composition, feed intake, feed efficiency, production traits, fertility, health, survival, longevity, and adaptation traits. For example, selection for increased growth may affect mature body size and feed requirements. Selection for higher milk production may influence fertility, health, body condition, and longevity. Selection for increased litter size may affect offspring survival and maternal performance.
- The relationship between production traits and fitness traits deserves particular attention. A breeding program that strongly emphasizes production without accounting for fertility, health, survival, and welfare may generate unfavorable correlated genetic changes. This is why modern breeding objectives increasingly incorporate a broader range of functional and fitness traits.
- Genetic correlations may also vary among populations, breeds, environments, management systems, and generations. A genetic relationship estimated in one population should not automatically be assumed to be identical in another. Differences in allele frequencies, selection history, measurement systems, management, and environmental conditions can influence genetic parameter estimates.
- Genotype–environment interaction can further complicate genetic relationships. The genetic correlation for a trait measured in two different environments can be interpreted as an indication of how consistently the same genetic factors influence performance across those environments. Low genetic correlation across environments can indicate that animals ranked highly in one environment may not rank similarly in another. This is particularly important for climate adaptation, heat tolerance, disease resistance, and low-input production systems.
- Genetic correlations can change over time as populations undergo selection. Strong directional selection can alter allele frequencies and genetic variances, potentially changing the genetic relationship between traits. Continuous monitoring of genetic parameters is therefore important in long-term breeding programs.
- Genetic correlations are also important for sustainable genetic improvement because they help breeders anticipate unintended consequences of selection. A breeding program should seek favorable improvement in economically important traits while avoiding excessive deterioration in fertility, health, welfare, longevity, adaptation, and other important characteristics. Balanced selection can therefore be more valuable than maximizing short-term response in a single trait.
- Genetic correlations also interact with genetic diversity and inbreeding. Intense selection concentrated on a small number of genetically superior animals can increase the use of popular sires and contribute to genetic concentration. Managing inbreeding, maintaining effective population size, and using strategies such as optimal contribution selection and mate allocation can help preserve genetic diversity while continuing genetic improvement.
- The interpretation of genetic correlations should therefore always consider biological meaning, statistical uncertainty, population structure, and the breeding objective. A high positive or negative estimate does not by itself indicate that selection should be increased or decreased. Breeders must consider whether the correlated change is desirable, economically important, biologically sustainable, and consistent with the overall breeding goal.
- Overall, genetic correlations between traits are a fundamental component of quantitative genetics and animal breeding. They describe the additive genetic relationships among traits and determine how selection on one trait can influence genetic change in other traits. By combining information on genetic covariance, heritability, breeding values, selection accuracy, multiple-trait evaluation, and genomic information, breeders can predict correlated responses and design more balanced breeding programs. Understanding genetic correlations is therefore essential for achieving efficient, sustainable, and biologically responsible genetic improvement across production, reproduction, health, welfare, survival, and adaptation traits.
