Exploiting the kernel trick to correlate fragment ions
for peptide identification via tandem mass spectrometry

Fu, Y., Yang, Q., Sun, R., Li, D., Zeng, R., Ling, C.X. and Gao, W.

Bioinformatics. Vol.20, pp1948-1954, 2004


Motivation: The correlation among fragment ions in tandem mass spectrum is crucial in reducing stochastic mismatches for peptide identification by database searching. Up to now, an efficient scoring algorithm that considers the correlative information in a tunable and comprehensive manner is still lacking in current systems. 

Results: This paper provides a promising approach to utilizing the correlative information to improve the peptide identification accuracy. The kernel trick, rooted in the statistical learning theory, is exploited to address this issue with low computational efforts. The common scoring method, tandem mass Spectral Dot Product (SDP), is extended to the Kernel SDP (KSDP). Experiments on a dataset previously reported demonstrate the effectiveness of the KSDP. The implementation on consecutive fragments shows a decrease of 10% in the error rate compared with the SDP. Our software tool, pFind, using a simple scoring function based on the KSDP, out-performs two SDP-based software tools SEQUEST and Sonar MS/MS in terms of identification accuracy.

Supplementary information


Ion trap spectra and the control DB can downloaded at http://regis.systemsbiology.net/mixdata/(Access authority from Andrew Keller is needed!).

Experiment: Comparison with other software tools:



This work was supported by the National Key Basic Research & Development Program (973) of China under Grant No. 2002CB713807. We thank Simin He, Yiqiang Chen, Hu Zhou and Xiaobiao Wang for valuable discussions. We acknowledge Dr. Andrew Keller for providing the MS/MS dataset, as well as the referees for helpful comments.

Contact: Yan Fu, yfu (at) ict.ac.cn,  http://www.jdl.ac.cn/user/yfu/index.htm