TY - JOUR T1 - A Two-State Epistasis Model Better Explains Complex Traits and Increases Power to Detect Additive Risk Variants JF - bioRxiv DO - 10.1101/017491 SP - 017491 AU - Kerry L. Bubb AU - Christine Queitsch Y1 - 2015/01/01 UR - http://biorxiv.org/content/early/2015/04/08/017491.abstract N2 - Despite more than a decade of effort, the genetic underpinnings of many complex traits and diseases remain largely elusive. It is increasingly recognized that a purely additive model, upon which most genome-wide association studies (GWAS) rely, is insufficient. Although thousands of significant trait-associated loci have been identified, they often explain little of the inferred genetic variance. Several factors have been invoked to explain the ‘missing heritability’, including epistasis. Accounting for all possible epistatic interactions is computationally complex and requires very large samples. Here, we propose a simple two-state epistasis model, in which individuals show either high or low variant penetrance with respect to a certain trait. The use of this model increases the power to detect additive trait-associated loci. We show that this model is consistent with current GWAS results and better fits heritability observations based on twin studies. We suggest that accounting for variant penetrance will significantly increase our power to identify underlying additive loci. ER -