Incorporating Prior Knowledge of Predictors into Penalized Classifiers with Multiple Penalty Terms

• Activity Number: 23
• Type: Contributed
• Date/Time: Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Feng Tai*+ and Wei Pan
• Companies: The University of Minnesota and The University of Minnesota
• Keywords: gene function ; PAM ; PPLS ; penalized classifier

Abstract

In the context of sample classifications with microarray data, many methods have been proposed. However, almost all the methods ignore existing biological knowledge and treat all the genes equally a priori. On the other hand, because some genes have been identified by previous studies to have biological functions or to be involved in pathways related to the outcome, incorporating this type of prior knowledge into a classifier can potentially improve both the predictive performance and interpretability of the resulting model. We propose a simple and general framework to incorporate such prior knowledge into building a penalized classifier. We group the genes according to their functional associations based on existing biological knowledge or data, and adopt group-specific penalty terms and penalization parameters. We apply the idea to PAM and PPLS and show the improvement.