Identifying biomarkers for acupuncture treatment via an optimization model

Abstract

Identifying biomarkers for acupuncture treatment is crucial to understand the mechanism of acupuncture effect at molecular level. In this study, we investigate the metabolic profiles of acupuncture treatment on several meridian points in human. To identify the subsets of metabolites that best characterize the acupuncture effect for each meridian point, a linear programming based model is proposed to identify biomarkers from the high-dimensional metabolic data. Specifically, we use nearest centroid as prototype to simultaneously minimize the number of selected features and leave-one-out cross validation error of the classifier. As a result, we reveal novel metabolite biomarkers for acupuncture treatment. Our result demonstrates that metabolic profiling might be a promising method to investigating the molecular mechanism of acupuncture. Comparison with other existing methods shows the efficiency and effectiveness of our new method. In addition, the method proposed in this paper is general and can be used in other high-dimensional applications, such as cancer genomics.