ACE model limitations
The ACE model is a statistical method used to analyse twin studies, separating the variation observed into three categories: additive genetic, common environment, and non-shared environment and error. It has several assumptions, however.1
One assumption of the ACE model is that a population has stochastic mating. The model assumes that individuals do not select others based on phenotypes.2
The model also assumes that different genotypes respond to environmental influence in the same ways. It also assumes that all genetic effects are additive.
Additive genetic effects occur where the expression of multiple genes contributes to a phenotype, and the resulting phenotype can be said to be the sum of these genetic factors. The ACE model does not allow dominance or epistasis to be taken into account, where the presence of some genes exert influence over others, potentially silencing their effect or otherwise modifying their action. Epistasis, for example, can lead to different effects together than on their own. For example, two individuals could have genes for blonde and brown hair respectively, but also carry the gene for baldness. The baldness gene would override the hair colour genes and prevent the growth of hair at some point (this is also non-additive, and the ACE model would fail with this example).3