Dissertation Defense Information
Department of Community Health, Biostatistics Section,
Center for Statistical Sciences, Brown University
Abstract: Growing interest on biological pathways has called for new statistical methods for modeling and testing the multi-dimensional pathway effect. In this talk, we propose a semiparametric logistic regression model for binary outcomes, where the clinical effects are modeled parametrically and the genetic pathway effect is modeled nonparametrically using kernel machines. The nonparametric function of a genetic pathway allows for the possibility that genes within the same pathway are likely to interact with each other and relate to the clinical outcome in a complicated way. We show that the kernel machine estimate can be formulated using a generalized linear mixed model. Estimation hence can proceed within the generalized linear mixed model framework using standard mixed model software. A score test based on a nonstationary Gaussian process is developed to test for the genetic pathway effect. The methods are illustrated using a prostate cancer data set and evaluated using simulations.
Brown Analysis Seminar
Scientific Computing Seminar
Department of Mathematics Colloquium
<--- 2007 Index