Computing P-Values for Peptide Identifications in Mass Spectrometry

Mass-spectrometry (MS) is a powerful experimental technology for ”sequencing” proteins in complex biological mixtures. Computational methods are essential for the interpretation of MS data, and a number of theoretical questions remain unresolved due to intrinsic complexity of the related algorithms. Here we design an analytical approach to estimate the confidence values of peptide identification in so-called database search methods. The approach explores properties of mass tags — sequences of mass values (m1 m2 ... mn), where individual mass values are distances between spectral lines. We define p-function — the probability of finding a random match between any given tag and a protein database — and verify the concept with extensive tag search experiments. We then discuss p-function properties, its applications for finding highly reliable matches in MS experiments, and a possibility to analytically evaluate properties of SEQUEST X-correlation function.