Defining epistasis

The term epistasis was initially used in the context of Mendelian inheritance; environmental effects are relatively unimportant for Mendelian traits, so individuals can be clearly assigned to one of a limited number of classes according to their phenotype. Here, epistasis was used to describe the situation in which the actions of one locus mask the allelic effects of another, in the same way that completely dominant alleles mask the effects of the recessive ones at the same locus. A clear example of this can be seen in (a) in which the dominant allele (I) at the KIT locus, which confers white-coat color in the pig, is dominant over all alleles at the MC1R locus (E), which confer a darker coat color. The effects of the various alleles at the E locus can only be determined in individuals with the recessive genotype ii at the I locus. This example was classically termed dominant epistasis, which gives a segregation ratio of 12:3:1 for white:black:brown, respectively. For complex traits, epistasis describes any interaction between two or more loci, such that the phenotype of any genotype cannot be predicted simply by summing the effects of individual loci. A fictive example with two loci with no epistasis for a complex trait is shown in (b). Here, the 3 lines for the effects of 3 genotypes at locus 1 run in parallel, indicating that the phenotypic effect is not influenced by the genotype at locus 2. Examples of epistasis for complex traits are shown in c-e. The first common pattern (c) is similar to Mendelian dominant epistasis shown in a, in which one locus in a dominant way suppresses the allelic effects of a second locus. In this example of growth in chickens, among-genotype variation for locus 2 is only expressed in the presence of the homozygous LL genotype at locus 1. Such epistasis often leads to individual quantitative trait loci (QTLs) having small average differences among genotypes and therefore not being detected unless epistasis is incorporated into the ana