Bayesian variable selection for binary outcomes in high-dimensional genomic studies using non-local priors
Open Access
- 6 January 2016
- journal article
- research article
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 32 (9) , 1338-1345
Abstract
Motivation: The advent of new genomic technologies has resulted in the production of massive data sets. Analyses of these data require new statistical and computational methods. In this article, we propose one such method that is useful in selecting explanatory variables for prediction of a binary response. Although this problem has recently been addressed using penalized likelihood methods, we adopt a Bayesian approach that utilizes a mixture of non-local prior densities and point masses on the binary regression coefficient vectors. Results: The resulting method, which we call iMOMLogit, provides improved performance in identifying true models and reducing estimation and prediction error in a number of simulation studies. More importantly, its application to several genomic datasets produces predictions that have high accuracy using far fewer explanatory variables than competing methods. We also describe a novel approach for setting prior hyperparameters by examining the total variation distance between the prior distributions on the regression parameters and the distribution of the maximum likelihood estimator under the null distribution. Finally, we describe a computational algorithm that can be used to implement iMOMLogit in ultrahigh-dimensional settings ( ) and provide diagnostics to assess the probability that this algorithm has identified the highest posterior probability model. Availability and implementation: Software to implement this method can be downloaded at: http://www.stat.tamu.edu/∼amir/code.html. Contact:wwang7@mdanderson.org or vjohnson@stat.tamu.edu Supplementary information: Supplementary data are available at Bioinformatics online.
Keywords
This publication has 28 references indexed in Scilit:
- Numbl inhibits glioma cell migration and invasion by suppressing TRAF5-mediated NF-κB activationMolecular Biology of the Cell, 2012
- Downregulation of SAV1 plays a role in pathogenesis of high-grade clear cell renal cell carcinomaBMC Cancer, 2011
- Bayes and empirical-Bayes multiplicity adjustment in the variable-selection problemThe Annals of Statistics, 2010
- Sure Independence Screening for Ultrahigh Dimensional Feature SpaceJournal of the Royal Statistical Society Series B: Statistical Methodology, 2008
- Mixtures of g Priors for Bayesian Variable SelectionJournal of the American Statistical Association, 2008
- The Dantzig selector: Statistical estimation when p is much larger than nThe Annals of Statistics, 2007
- The Adaptive Lasso and Its Oracle PropertiesJournal of the American Statistical Association, 2006
- A Coupling-Regeneration Scheme for Diagnosing Convergence in Markov Chain Monte Carlo AlgorithmsJournal of the American Statistical Association, 1998
- Studying Convergence of Markov Chain Monte Carlo Algorithms Using Coupled Sample PathsJournal of the American Statistical Association, 1996
- Studying Convergence of Markov Chain Monte Carlo Algorithms Using Coupled Sample PathsJournal of the American Statistical Association, 1996