Advantages of the Bagidis methodology for metabonomics analyses: application to a spectroscopic study of Age-related Macular Degeneration

The Bagidis methodology proposes a distance measure between spectra, that takes into account, in a unified framework, both horizontal shifts and amplitudes variations that might affect spectral peaks. The method relies on the expansion of the spectra in unbalanced Haar wavelet bases. Its opportunity for investigating 1HNMR spectra in metabonomics is illustrated here in the framework of a study of an eye disease: age-related macular degeneration. Visual analysis, disease detection model and search for biomarkers are proposed here and compared with known methods.