DNA microarray technology enables us to monitor the expression levels of thousands of genes simultaneously and has emerged as a promising tool for disease diagnosis. We present a review of recent developments in Bayesian statistical methods for microarray data. In particular, we focus on Bayesian gene selection for and survival analysis. Owing to the large number of genes and the complexity of the data, we use Markov chain Monte Carlo (MCMC) based stochastic search algorithms for inference. Other recent technical developments in Bayesian modeling for microarray data are summarized. The methodology is illustrated using several well-analyzed cancer microarray datasets. © 2005 Elsevier B.V. All rights reserved.