Multi-Label Classifier Design for Predicting the Functional Types of Antimicrobial Peptides

Abstract:

t has special meaning for drug design as well as basic research to study Antimicrobial peptides (AMPs) because they have been demonstrated to kill Gram negative and Gram positive bacteria, mycobacteria, enveloped viruses, fungi and even transformed or cancerous cells. In view of this, it is highly desired to develop an effective computational method for accurately predicting the functional types of AMPs because it can provide us with more candidates and useful insights for drug design. AMP functional recognition is in fact a multi-label classification problem. In this study, up to six kinds of physicochemical properties value are selected to code the AMP sequence as physical-chemical property matrix (PCM), and then auto and cross covariance transformation is performed to extract features from the PCM for AMP sequence expression; At last, a clever use of Fuzzy K nearest neighbor rule will help identify the multiple functions of a query AMP. As a result, the overall classification accuracy about 65% has been achieved through the rigorous Jackknife test on a newly constructed benchmark AMP dataset.

• Periodical: Advanced Materials Research (Volumes 718-720)
• Pages: 293-298
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.718-720.293
• Online since: July 2013
• Authors: Pu Wang, Xuan Xiao
• Keywords: Antimicrobial Peptide, Jackknife Test, Multi-Label Classification, Physical-Chemical Property Matrix

References

[1] Z. Wang, G. Wang, APD: the Antimicrobial Peptide Database, Nucleic Acids Res, 32(2004) D590-592.

DOI: 10.1093/nar/gkh025

[2] C.D. Fjell, R.E. Hancock, A. Cherkasov A, AMPer: a database and an automated discovery tool for antimicrobial peptides, Bioinformatics, 23(2007)1148–1155.

DOI: 10.1093/bioinformatics/btm068

[3] S. Lata, B.K. Sharma, G.P. Raghava, Analysis and prediction of antibacterial peptides, BMC Bioinformatics, 8(2007)263.

DOI: 10.1186/1471-2105-8-263

[4] S. Lata, N.K. Mishra, G.P. Raghava, AntiBP2: improved version of antibacterial peptide prediction, BMC Bioinformatics, 11(2010)19-7.

DOI: 10.1186/1471-2105-11-s1-s19

[5] H.B. Shen, K.C. Chou, PseAAC: A flexible web server for generating various kinds of protein pseudo amino acid composition, Analytical Biochemistry, 373(2008)386-388.

DOI: 10.1016/j.ab.2007.10.012

[6] M. Lapinsh, A. Gutcait, et al., Classification of G-protein coupled receptors by alignment -independent extraction of principal chemical properties of primary amino acid sequences, Protein Science, 11(2002)795-805.

DOI: 10.1110/ps.2500102

[7] Y. Guo, M. Li, et al., Predicting G-protein coupled receptors–G-protein coupling specificity based on autocross-covariance transform, Proteins: Structure, Function, and Bioinformatics, 65(2006)55-60.

DOI: 10.1002/prot.21097

[8] Q. Dong, S. Zhou, et al., A new taxonomy-based protein fold recognition approach based on autocross-covariance transformation, Bioinformatics, 25(2009)2655-2662.

DOI: 10.1093/bioinformatics/btp500

[9] J.M. Keller, M.R. Gray, et al., A frezzy K-nearest neighbor algorithm, IEEE transactions on systems, man, and cybernetics, 15(1985)580-5.

DOI: 10.1109/tsmc.1985.6313426

[10] X. Shen, M. Boutell, J. Luo, C. Brown, Multi-label machine learning and its application to semantic scene classification, Proceedings of the 2004 International Symposium on Electronic Imaging. San Jose, California, USA, 2004, pp.18-22.