In this article, we discuss issues that arise in the statistical analysis of complex genetic traits using inbred line crosses. Current methods of analysis that are in widespread use often fail to detect gene x gene interactions and thus can present an overly simplified view of a complex genetic system. Uncertainty regarding estimated parameters and model specification is inherent in any statistical analysis of data. Our goal is to fully determine the component genes and their potential interactions based on information derived from an experimental cross. It is important for statisticians and geneticists alike to recognize both the power and the limitations of this approach.
/lp/association-for-computing-machinery/complexity-and-uncertainty-in-genetics-hAvSKzH0g2