Hypothesis testing for the genetic background of quantitative traits

The testing of Bayesian point null hypotheses on variance component models have resulted in a tough assignment for which no clear and generally accept ed method exists. In this work we present what we believe is a succeeding a pproach to such a task. It is based on a simple reparameterization of the m odel in terms of the total variance and the proportion of the additive gene tic variance with respect to it, as well as on the explicit inclusion on th e prior probability of a discrete component at origin. The reparameterizati on was used to bypass an arbitrariness related to the impropriety of uninfo rmative priors onto unbounded variables while the discrete component was ne cessary to overcome the zero probability assigned to sets of null measure b y the usual continuous variable models. The method was tested against compu ter simulations with appealing results.