A modeling of complex trait phenotypic variance determinants

Abstract

Studies have now shown that the heritability of some complex traits, such as human height, can be virtually fully captured via potential use of sufficiently powered approaches that can characterize the associated collective common- and rare-variant additive genetic architecture. However, for other traits, including complex disease traits, full recovery of such narrow sense heritability would still likely fall far short of respective heritability estimates yielded from pedigree-based analyses such as twin studies. Here, it is proposed that such traits could also involve additional types of relevant architecture and underlying genetic mechanism, such that interaction of somatic variants with heritable variants may represent an underappreciated component. The theoretical model suggested predicts that some relevant heritability estimates are systematically inflated by twin studies, and that instead a significant proportion of the phenotypic variances may be explained by specialized types of heritable genotype-by-environment interaction.