We present in this paper a novel fragmentation event model for peptide identification by tandem mass spectrometry. Most current peptide identification techniques suffer from the inaccuracies in the predicted theoretical spectrum, which is due to insufficient understanding of the ion generation process, especially the b/y ratio puzzle. To overcome this difficulty, we propose a novel fragmentation event model, which is based on the abundance of fragmentation events rather than ion intensities. Experimental results demonstrate that this model helps improve database searching methods. On LTQ data set, when we control the false-positive rate to be 5%, our fragmentation event model has a significantly higher true positive rate (0.83) than SEQUEST (0.73). Comparison with Mascot exhibits similar results, which means that our model can effectively identify the false positive peptide-spectrum pairs reported by SEQUEST and Mascot. This fragmentation event model can also be used to solve the problem of missing peak encountered by De Novo methods. To our knowledge, this is the first time the fragmentation preference for peptide bonds is used to overcome the missing-peak difficulty. Availability: http://www.bioinfo.org.cn/MSMS/. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Lin, Y., Qiao, Y., Sun, S., Yu, C., Dong, G., & Bu, D. (2008). A fragmentation event model for peptide identification by mass spectrometry. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4955 LNBI, pp. 154–166). https://doi.org/10.1007/978-3-540-78839-3_14
Mendeley helps you to discover research relevant for your work.