ďCharacterizing the contribution of genetic variations to

the prediction of human disease risk "

November 9, 2010
A Presentation by Greg Dyson

The augmented Patient Rule Induction Method (PRIM) developed by Dyson et al is a novel model building strategy for evaluating an individualís risk which acknowledges the etiological heterogeneity of a disease. It is designed to identify combinations of risk factor values that characterize mutually exclusive subgroups of individuals that differ in average risk as measured by the cumulative incidence of the disease of interest. This analytical strategy addresses the question of which alterations in which risk factors best predict the disease of interest in which subset of individuals of the population at large from which the sample under study was drawn. In this presentation, we employ a stepwise application of the PRIM model building strategy to test the hypothesis that information about 5 apolipoprotein E (APOE) genotypes significantly improves the prediction of ischemic heart disease (IHD) in particular sub-samples of individuals characterized by selected subsets of values of the traditional risk factors.

Greg Dyson is an Assistant Professor in the Biostatistics Core of the Karmanos Cancer Institute. He obtained a PhD in Statistics from the University of Michigan. Prior his current position, he worked as a Research Investigator in the Human Genetics Department at the University of Michigan. His research interests include statistical genetics, bioinformatics, experimental design and statistical computing.