Bayesian Model Averaging: Development of an Improved Multi-Class, Gene Selection and Classification Tool for Microarray Data

Abstract

Accurate classification of samples using gene expression profiles is critically dependent on the method used to select relevant genes. We present the Bayesian Model Averaging (BMA) method for gene selection and classification of microarray data. Typical gene selection and classification procedures ignore model uncertainty and use a single set of relevant genes (model) to predict the class. BMA accounts for the uncertainty about the best set to choose by averaging over multiple models (sets of potentially overlapping relevant genes). We showed that BMA selects smaller numbers of relevant genes (compared to other methods) and achieves high prediction accuracy on three microarray datasets. Our BMA algorithm is applicable to microarray datasets with any number of classes, and outputs posterior probabilities for the selected genes and models. Our selected models typically consist of only a few genes. The combination of high accuracy, small numbers of genes and posterior probabilities for the predictions, should make BMA a powerful tool for developing diagnostics from expression data.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 08, 2004
Accession Number
ADA454826

Entities

People

  • Adrian Raftery
  • Ka Y. Yeung
  • Roger E. Bumgarner

Organizations

  • University of Washington

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Software
  • Bayesian Networks
  • Breast Cancer
  • Classification
  • Computational Science
  • Computer Programs
  • Data Mining
  • Dna Microarrays
  • Information Science
  • Lymphatic Diseases
  • Lymphocytes
  • Machine Learning
  • Monte Carlo Method
  • Probability
  • Supervised Machine Learning
  • Test Sets

Fields of Study

  • Biology

Readers

  • Computational Modeling and Simulation
  • Molecular and genetic basis of cancer.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks