Computational aspects of protein identification by mass spectrometry.

Recent developments in proteomics and genomics provide huge quantities of data to analyze. Automatic interpretation of mass spectrometry data has become essential for high-throughput processes aiming to study complete proteomes. There exist two main sources of mass spectrometric data: peptide mass fingerprint and fragmentation spectra, both of which require specific bioinformatic algorithms. We present a survey of these algorithms and discuss the efficiency of the different approaches and the possible improvements that may lead to a complete automatic high-throughput identification process.