Mining a Tandem Mass Spectrometry Database to Determine the Trends and Global Factors Influencing Peptide Fragmentation

Monday, September 29, 2003 - 9:30am - 10:20am
Keller 3-180
Terence Speed (University of California, Berkeley)
A statistical and non-statistical method have been used to analyse the gas phase fragmentation behavior of protonated peptides that involves mining a database of several thousand unique product ion spectra derived from tryptic digestion and low-energy collision induced dissociation in a quadrupole ion trap mass spectrometer. This bioinformatic approach has resulted in the derivation of a ³relative proton mobility scale² that takes into account both the charge state and the amino acid composition of a peptide, and provides an effective classification system for categorizing peptide MS/MS spectra for subsequent data mining and statistical analysis. We show that the most important factor influencing fragmentation is proton mobility and that peptides classified as non-mobile generally give scores below currently acceptable thresholds using current MS/MS search algorithms. An amino acid residues preference for N- and/or C-terminal cleavage has been quantified in accordance with the proton mobility scale and the trends determined are predictable based on an analysis of the most abundant cleavage sites. (Joint work with Eugene A. Kapp, Frédéric Schütz and Richard J. Simpson)