- Epistatic Variance is a component of genetic variance that arises from interactions between genetic effects at different loci. It is an important concept in quantitative genetics because the effect of an allele at one genetic locus may depend on the genotype present at another locus. These interactions can contribute substantially to differences in complex traits and influence inheritance, natural selection, breeding, and evolution.
- Epistasis refers broadly to interactions between genes or genetic loci in which the effect of one locus depends on variation at another locus. Epistatic variance describes the contribution of these interactions to phenotypic differences among individuals in a population.
- In a simplified decomposition of phenotypic variance, genetic variance can be divided into additive genetic variance, dominance variance, and epistatic or interaction variance. A simplified expression is VP=VA+VD+VI+VEV_P = V_A + V_D + V_I + V_E, where VAV_A is additive genetic variance, VDV_D is dominance variance, VIV_I represents interaction variance, and VEV_E represents environmental variance. Real biological systems can be more complex and may include additional covariance and interaction components.
- The most important distinction is between additive effects and epistatic effects. An additive genetic model assumes that the contributions of alleles can be summed to estimate their combined genetic effect. Epistasis occurs when the combined effect of variants at different loci differs from what would be expected from their individual effects.
- For example, suppose a variant at locus A affects a biological pathway and another variant at locus B affects a different component of the same pathway. The effect of the allele at locus A may be different depending on which allele is present at locus B. The resulting interaction represents an epistatic effect.
- Epistatic interactions can occur between many different types of genetic variants. They may involve protein-coding genes, regulatory regions, transcription factors, signaling pathways, metabolic genes, or other functional elements. Because biological systems contain interconnected molecular networks, interactions between genetic loci are common in principle even though detecting them statistically can be challenging.
- Epistasis can be classified into different forms depending on the genetic model. Functional epistasis refers to biological interactions between gene products or pathways, while statistical epistasis refers to deviations from a specified statistical model of genetic effects. These concepts are related but should not be treated as identical.
- Several classical patterns of epistasis have been described in Mendelian genetics. These include recessive epistasis, dominant epistasis, duplicate gene interactions, complementary gene action, and other modified segregation patterns. In quantitative genetics, however, epistasis is usually considered more broadly as statistical interaction among loci contributing to variation in a quantitative trait.
- Epistatic variance is influenced by allele frequencies. The contribution of an interaction to population-level genetic variance depends not only on the magnitude of the interaction but also on how frequently the relevant alleles occur. Changes in allele frequencies can therefore increase or decrease epistatic variance without changing the underlying biological interaction.
- Genetic background is also important. The effect of a particular allele can depend on the other variants carried by an individual. This phenomenon is sometimes described as genetic background dependence and is one reason why the same mutation can produce different phenotypic effects in different genetic backgrounds.
- Epistatic variance is distinct from dominance variance. Dominance describes interactions between alleles at the same locus, whereas epistasis generally describes interactions between genetic effects at different loci. Both are forms of non-additive genetic variation, but they arise from different types of genetic relationships.
- Epistatic variance is also distinct from additive genetic variance. Additive effects are especially important because they contribute predictably to resemblance between parents and offspring. Epistatic interactions can be transmitted through offspring, but the specific combinations of alleles at different loci are reshuffled by meiosis, recombination, and Mendelian segregation.
- Because epistatic combinations can be broken apart through recombination, the contribution of epistatic variance to the response to selection can be more difficult to predict than additive genetic variance. Nevertheless, epistasis can influence short-term and long-term evolutionary change by affecting fitness and the genetic architecture of complex traits.
- The relationship between epistasis and heritability is important. Narrow-sense heritability is based on additive genetic variance:
- h2=VA/VPh^2 = V_A / V_P
- Epistatic variance is not directly included in narrow-sense heritability. Broad-sense heritability, in contrast, considers total genetic variance and can therefore include additive, dominance, and epistatic components:
- H2=VG/VPH^2 = V_G / V_P
- The distinction matters because a trait can have substantial genetic influence while showing relatively little narrow-sense heritability if a large proportion of its genetic variance arises from non-additive interactions.
- Epistasis can influence the breeding value of an individual. Breeding value primarily represents the additive genetic contribution expected to be transmitted to offspring, whereas an individual’s total genotypic value can also include dominance and epistatic effects. Therefore, two individuals with similar breeding values may differ in their total genotypic values because of different genetic interactions.
- Epistatic variance is particularly relevant to genetic architecture. Complex traits are often controlled by many loci, and these loci may interact with one another. The genetic architecture of a trait can therefore include additive effects, dominance effects, epistatic interactions, pleiotropy, linkage, and linkage disequilibrium.
- Epistasis can also contribute to the genetic basis of polygenic traits. A polygenic trait may be influenced by hundreds or thousands of variants, and interactions among subsets of these variants can modify their combined effects. This means that the phenotype of a complex trait cannot always be accurately predicted by simply adding the independent effects of individual genetic variants.
- In evolutionary genetics, epistasis can influence how populations respond to natural selection. The fitness effect of an allele may depend on the genetic background in which it occurs. Consequently, an allele that is advantageous in one population or genetic background may have a weaker effect, no effect, or even a disadvantageous effect in another.
- Epistasis can contribute to the formation of fitness landscapes, which describe how combinations of genetic variants relate to biological fitness. In a purely additive landscape, the effects of genetic changes can be predicted by adding their individual contributions. With epistasis, the fitness effect of one mutation can depend on which other mutations are already present.
- This can create genetic constraints and evolutionary pathways. Some combinations of mutations may be beneficial only when particular background variants are present. Other combinations may be harmful even though each individual mutation appears beneficial when considered separately.
- Epistasis is also important in understanding compensatory mutations. A mutation that initially reduces biological function may be followed by another mutation that partially or completely restores function. The interaction between the two variants represents an example of genetic dependence across loci.
- The importance of epistasis is particularly clear in biological pathways. Genes rarely operate completely independently. Proteins interact with other proteins, enzymes participate in metabolic pathways, and regulatory genes influence the expression of other genes. These relationships provide biological mechanisms through which genetic interactions can arise.
- Epistasis can also interact with gene–environment interaction (G×E). The interaction between two genetic loci may differ depending on environmental conditions. This can produce higher-order interactions in which genotype combinations respond differently across environments.
- For example, one combination of alleles may provide high performance under favorable environmental conditions but perform poorly under environmental stress, while another genetic combination may show the opposite pattern. Such relationships complicate the prediction of phenotypes from genetic information alone.
- Epistatic variance is relevant to plant and animal breeding because interactions between loci can influence economically important traits such as growth, fertility, yield, disease resistance, and stress tolerance. Breeding programs may exploit favorable genetic combinations, although maintaining specific combinations across generations can be difficult because recombination reshuffles alleles.
- Hybrid breeding can also involve epistatic effects. Hybrid performance may result from combinations of dominance and epistatic interactions between parental genomes. This is one reason why specific crosses can sometimes outperform the performance predicted from parental lines alone.
- Modern genomic selection provides opportunities to model non-additive genetic effects. Traditional genomic prediction models often emphasize additive effects because they are easier to estimate and tend to provide strong predictive performance. However, models can also incorporate dominance and epistatic interactions when sufficient data are available.
- Detecting epistatic interactions statistically is challenging because the number of possible interactions increases rapidly with the number of genetic markers. If a genome contains a very large number of variants, testing every possible pair or higher-order combination can produce enormous numbers of statistical comparisons.
- Genome-wide association studies (GWAS) can therefore identify single-variant associations more easily than they can detect genome-wide interactions. Specialized statistical methods, larger sample sizes, biological knowledge, and carefully controlled multiple-testing procedures are often required to identify reliable epistatic effects.
- QTL mapping can also be used to investigate interactions between quantitative trait loci. Researchers may test whether the effect of one QTL depends on the genotype at another QTL. Such analyses can reveal genetic interactions that would be missed by considering each locus independently.
- The interpretation of statistical epistasis requires caution. A statistical interaction does not necessarily prove a direct physical interaction between the proteins encoded by two genes. Statistical epistasis can arise from biological pathway interactions, population structure, allele frequencies, model assumptions, linkage, or other features of genetic architecture.
- Epistatic variance can also change during evolution. As allele frequencies change through natural selection, genetic drift, mutation, and gene flow, the contribution of particular genetic interactions to population-level variance may change. Therefore, the amount of epistatic variance is not necessarily constant across populations or generations.
- Recombination is especially important because it changes the combinations of alleles at different loci. Recombination can break apart favorable or unfavorable genetic combinations and create new combinations. This can alter the expression and contribution of epistatic interactions over time.
- Linkage disequilibrium (LD) can also influence estimates of epistatic effects. When alleles at different loci are associated non-randomly, their effects may be correlated in a population. Distinguishing true biological interactions from statistical associations caused by LD can therefore be difficult.
- Epistasis is relevant to human genetics and the study of complex diseases. Many diseases and traits involve multiple biological pathways, and genetic variants may interact to influence disease risk or treatment response. However, reliable detection of epistatic effects in human populations generally requires large datasets and careful replication because interaction effects are often small relative to the number of possible genetic combinations.
- Epistatic interactions may also influence pharmacogenomics, where genetic variation affects responses to medications. The effect of one genetic variant on drug metabolism or treatment response may depend on other variants, although many such interactions remain difficult to identify and validate.
- In conservation and evolutionary genetics, epistatic variance can contribute to a population’s capacity for adaptation. Populations with diverse combinations of genetic variants may contain genetic interactions that become advantageous when environmental conditions change. However, small populations may lose important genetic combinations through genetic drift and inbreeding.
- Epistatic variance is therefore an important but complex component of genetic variation. It demonstrates that the effects of genes cannot always be understood independently and that genetic architecture can involve extensive interactions among loci.
- Understanding epistatic variance helps connect genetic variation, additive genetic variance, dominance variance, quantitative genetics, genetic architecture, heritability, breeding value, genomic selection, and evolutionary genetics. It provides a framework for studying how interactions among different genetic loci contribute to phenotypic differences and influence the response of populations to selection and environmental change.
- Epistatic variance ultimately highlights one of the central principles of modern genetics: the effect of a genetic variant depends not only on the variant itself but also on the broader genetic context in which it occurs. This context can influence phenotype, fitness, inheritance, breeding outcomes, and evolutionary change.
