Computational Methods for the Discovery of Metabolic Markers of Complex Traits.

摘要: Metabolomics uses quantitative analyses of metabolites from tissues or bodily fluids to acquire a functional readout the physiological state. Complex diseases arise influence multiple factors, such as genetics, environment and lifestyle. Since genes, RNAs proteins converge onto terminal downstream metabolome, metabolomics datasets offer rich source information in complex convoluted presentation. Thus, powerful computational methods capable deciphering effects many upstream influences have become increasingly necessary. In this review, workflow metabolic marker discovery is outlined metabolite extraction model interpretation validation. Additionally, current research various disease areas examined identify gaps trends use several statistical algorithms. Then, we highlight discuss three advanced machine-learning algorithms, specifically ensemble learning, artificial neural networks, genetic programming, that are currently less visible, but budding with high potential for utility research. With an upward trend highly-accurate, multivariate models literature, diagnostic biomarker panels more recently achieving accuracies approaching exceeding traditional procedures. This review aims provide overview promote up-to-date by researchers.