The information encrypted in accurate peptide masses-improved protein identification and assistance in glycopeptide identification and characterization.

Analytically useful information from accurate mass data for peptides with an error of </=20 ppm is discussed. The deltamass (= mass value following the decimal point) distribution of natural peptides is extracted from a protein database. Compared with the random peptide data, the natural data show a higher average deltamass value and a smaller width of the mass distribution. This deviation can be ascribed to the non-random abundances of the standard amino acids. In particular, accurate mass data for peptides located near the edges of the natural mass distribution contain analytical information. Mass data near the edges generate very few hits in a protein database search and are therefore highly specific for protein identification. Mass signals near the low-mass edge indicate either a high probability that the peptide contains one or several cysteine sites, or that the peptide is highly acidic due to the presence of several D and/or E residues or that it is a glycopeptide. Mass data near the high-mass edge indicate a non-polar peptide with a high abundance of the non-polar amino acids leucine, isoleucine and valine. An Internet page is introduced that analyzes the deviation of a peptide mass from the average deltamass value and that supports the characterization of glycopeptides found near the low-mass edge of the mass distribution. Copyright 2000 John Wiley & Sons, Ltd.