Search

The identification of the minimum effective dose is of high importance in the drug development process. In early stage screening experiments, establishing the minimum effective dose can be translated into a model selection based on information criteria. The presented alternative, Bayesian variable selection approach, allows for selection of the minimum effective dose, while taking into account model uncertainty. The performance of Bayesian variable selection is compared with the generalized order restricted information criterion on two dose-response experiments and through the simulations study. Which method has performed better depends on the complexity of the underlying model and the effect size relative to noise. Martin Otava and Ziv Shkedy gratefully acknowledge the support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy). Martin Otava gratefully acknowledge the financial support of the Research Project BOF11DOC09 of Hasselt University.

Recently, preclinical microarray experiments have become increasingly common laboratory tools to investigate the activity of thousands of genes simultaneously and their response to a certain treatment (Amaratunga and Cabrera 2004). In some experiments, in addition to the gene expressions, other responses are also available. In such situations, the primary question of interest is to identify whether or not the gene expressions can serve as biomarkers for the responses. In addition to gene expressions, metabolites are potential biomarkers for some responses as well. In the present study, we focus on the identification of genomic biomarkers, based on gene and metabolite expression for depression. One measure of the level of depression is the Hamilton Depression Scale (HDS or HAMD) which is a test measuring the severity of depressive symptoms in individuals. The data for this study are a result of a clinical trial in which both HAMD and gene/metabolites expression were measured. We use three modeling approaches commonly used in the surrogate marker validation theory to select and evaluate a set of genes and metabolites as possible biomarkers for depression, as measured by the HAMD score. In addition to gene and metabolite specific biomarkers, we use supervised principal components analysis and supervised partial least squares regression technique to construct a joint biomarker that uses information from all genes/metabolites in the array. The authors gratefully acknowledge support from IAP research Network P6/03 of the Belgian Government (Belgian Science Policy).