Metabolomics for the future of personalized medicine through information and communication technologies.

Abstract

Although we have empirically found many ways to improve our lives, the challenge of cur­ ing diseases and extending our lifespan poses enormous problems for which we have, at the moment, no adequate answers. For instance, in spite of expenditures estimated to be on the order of €200 billion for cancer research over the last few decades, the definitive cure rates of most common forms of cancer have hardly increased. We will, therefore, need to take both the genetic and environmental effects differing between individuals much more into account. This will not only require a much more detailed characterization of each individual patient but will also require a new generation of ana lysis tools. We will need to convert much larger amounts of data into predictions on the effects of different therapies/prevention strategies. In turn, this will need to rely on a combination of mathematical and probabilistic modeling and advanced statistical inferential methodologies such as multivariate statistics and machine learn­ ing. Therefore, current empirical approaches will have to be replaced (or extended) by compu­ tational models adequate to take into account the complexity of the whole human being. A futuristic, data­intensive strategy to imple­ ment these ideas in the personalized medicine of the future would be the development of an individual model that, gathering together all the data available from the ’omic sciences, will build a ‘virtual patient’ associated to each individual. In this context, metabolomics will have a particular role with respect to the other ’omic sciences, because of its ability to detect, in real time, the adaptive multiparametric response of the organisms to pathophysiological stimuli or genetic modifications [1].