Provides a method that supports fully automated and quantitative nuclear magnetic resonance (NMR)-based metabolomics of complex mixtures. Bayesil was developed to divide the spectrum into small blocks and represents the sparse dependencies between these blocks. It then performs approximate inference over this model as a surrogate for spectral profiling, yielding the most probable metabolic profile.
Department of Computing Science, University of Alberta, Edmonton, AB, Canada; Alberta Innovates Center for Machine Learning, Edmonton, AB, Canada; Department of Biological Sciences, University of Alberta, Edmonton, AB, Canada; Fiorgen Foundation, Florence, Italy; Centro Risonanze Magnetiche, University of Florence, Florence, Italy; National Research Council, National Institute for Nanotechnology, Edmonton, AB, Canada
Bayesil funding source(s)
Supported by Alberta Innovates–Health Solutions, Alberta/ Pfizer Translational Research Fund, Metabolomics Innovation Centre, Natural Sciences and Engineering Research Council of Canada, Canadian Institutes of Health Research, Alberta Innovates Technology Futures, Queen Elizabeth II graduate scholarships, Alberta Innovates Centre for Machine Learning, CIHR grant reference number 111062.