Using multivariate curve resolution to improve proteomic mass spectra classification

This article describes a novel proteomic pattern analysis algorithm for biomarker discovery using MALDI-TOF or SELDI-TOF mass spectrometry. The algorithm (MCR-marker) is based on the combination of Multivariate Curve Resolution with classification methods for the detection of potential biomarkers. Precise m/z values or m/z bins have been widely used as proteomics biomarkers or discriminatory patterns to classify biological samples in various cancer studies. However, the lack of reproducibility of these discriminatory patterns has been a major criticism and will be a major hurdle before it can be a useful clinical tool. The MCR-marker algorithm applies singular value decomposition to select differentially expressed m/z windows. In each selected m/z window, potential biomarkers are identified from MCR-resolved peak profiles that show better performance than the precise m/z values or m/z bins. The identified potential biomarkers are not dependent on the selection of MCR methods and consist of clearly detectable peaks, which may represent identifiable proteins, protein fragments or peptides. The algorithm is validated on two data sets from the literature.