|
Abstract:
|
We propose a Bayes regression model to study the time course expression profiles of a virus gene, based on data from micro-array experiments. Since the expression level of a virus gene in a cell is constantly zero initially, increasing for a while, and then decreasing, we consider a regression model in which the mean function takes this shape restriction, with the prior introduced by Bernstein polynomials, which incorporates this geometric information naturally. We implement the inference by Metropolis-Hastings reversible-jump algorithm. We will report simulation results to indicate the performance of this method and illustrate it with the analysis of a real data set for Baculavirus.
|
