Predictive Modeling in Proteomics-based Disease Detection

Recent advent of mass-spectrometry data generated by proteomic technology provides a new type of biological information which is very promising in the search for diagnostic and therapeutic approaches that enables the early detection of fatal diseases and the development of personalized medicine. Successful analysis of such high-throughput proteomic data relies much on signal-processing and pattern-recognition techniques. This paper addresses the application of prediction models for cancer detection using mass spectral data.