Bayesian Variable Selection in Multinomial Probit Models to Identify Molecular Signatures of Disease Stage
- 27 August 2004
- journal article
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 60 (3) , 812-819
Abstract
Summary Here we focus on discrimination problems where the number of predictors substantially exceeds the sample size and we propose a Bayesian variable selection approach to multinomial probit models. Our method makes use of mixture priors and Markov chain Monte Carlo techniques to select sets of variables that differ among the classes. We apply our methodology to a problem in functional genomics using gene expression profiling data. The aim of the analysis is to identify molecular signatures that characterize two different stages of rheumatoid arthritis.Keywords
This publication has 25 references indexed in Scilit:
- Optimal predictive model selectionThe Annals of Statistics, 2004
- Gene selection in arthritis classification with large‐scale microarray expression profilesComparative and Functional Genomics, 2003
- Bayes Model Averaging with Selection of RegressorsJournal of the Royal Statistical Society Series B: Statistical Methodology, 2002
- Bayesian Models for Gene Expression With DNA Microarray DataJournal of the American Statistical Association, 2002
- Life and death in the JUNgleTrends in Molecular Medicine, 2001
- Paxillin Localizes to the Lymphocyte Microtubule Organizing Center and Associates with the Microtubule CytoskeletonJournal of Biological Chemistry, 2000
- Model Selection and Accounting for Model Uncertainty in Graphical Models Using Occam's WindowJournal of the American Statistical Association, 1994
- Bayesian Analysis of Binary and Polychotomous Response DataJournal of the American Statistical Association, 1993
- The Calculation of Posterior Distributions by Data AugmentationJournal of the American Statistical Association, 1987
- Ridge Regression: Biased Estimation for Nonorthogonal ProblemsTechnometrics, 1970