Risk Classification with an Adaptive Naive Bayes Kernel Machine Model

• Activity Number: 227
• Type: Topic Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Jessica Minnier*+ and Tianxi Cai and Jun Liu
• Companies: Harvard University and Harvard University and Harvard University
• Keywords: Risk prediction ; Genetic association ; Kernel PCA ; Genetic pathways ; Gene-set analysis ; Kernel Machine Regression

Abstract

The complex genetic architecture of disease makes it difficult to identify genomic markers associated with disease risk. Standard risk prediction models often rely on additive or marginal relationships of markers and the phenotype of interest. These models perform poorly when associations involve interactions and non-linear effects. We propose a multi-stage method relating markers to disease risk by first forming gene-sets based on biological criteria. With a naive bayes kernel machine model, we estimate gene-set specific risk models that relate each gene-set to the outcome. Second, we aggregate across gene-sets by adaptively estimating weights for each set. The KM framework models the potentially non-linear effects of predictors without specifying a particular functional form. Estimation and predictive accuracy are improved with kernel PCA in the first stage and adaptive regularization in the second stage to remove non-informative regions from the final model. Prediction accuracy is assessed with bias-corrected ROC curves and AUC statistics. Numerical studies suggest that the model performs well in the presence of non-informative regions and both linear and non-linear effects.