Computing P-values for peptide identifications in mass spectrometry

摘要: Mass-spectrometry (MS) is a powerful experimental technology for"sequencing" proteins in complex biological mixtures. Computational methodsare essential for the interpretation of MS data, and number theoretical questionsremain unresolved due to intrinsic complexity related algorithms.Here we design an analytical approach estimate confidence values peptideidentification so-called database search methods. The exploresproperties mass tags -- sequences (m1 m2...mn), where individualmass are distances between spectral lines. We define p-function probability finding random match any given tag proteindatabase verify concept with extensive experiments.We then discuss properties, its applications highly reliablematches experiments, possibility analytically evaluateproperties SEQUEST X-correlation function.