Title: On MEBID, an Integrated Multivariate Method for Efficient Biomarker Discovery
Abstract:
We will present the framework of MEBID, an integrated multivariate method for efficient biomarker discovery. Biomarkers, or biological markers, are quantifiable characteristics of biological processes. In medical settings, they can be used as diagnostic or prognostic tools. They may permit to investigate and understand the mechanism of development of pathology, and are thus important in drug discovery and drug development. The methodology is built upon the envelope regression methodologies of RD Cook. Envelope regression methods constitute a novel multivariate tool that allows an efficient estimation of regression parameters, taking into consideration the correlation structure of the data.