Machine Learning Techniques: Approach for Mapping of MHC Class Binding Nonamers

Gomase V.S.¹*, Yash Parekh², Subin Koshy³, Siddhesh Lakhan^4
1Department of Bioinformatics, Padmashree Dr. D.Y. Patil University, Navi Mumbai, 400614, India
²Department of Bioinformatics, Padmashree Dr. D.Y. Patil University, Navi Mumbai, 400614, India
³Department of Bioinformatics, Padmashree Dr. D.Y. Patil University, Navi Mumbai, 400614, India
⁴Department of Bioinformatics, Padmashree Dr. D.Y. Patil University, Navi Mumbai, 400614, India
* Corresponding Author : virusgene1@yahoo.co.in

Received : -     Accepted : -     Published : 15-06-2009
Volume : 1     Issue : 1       Pages : 1 - 5
Genetics 1.1 (2009):1-5
DOI : http://dx.doi.org/10.9735/0975-2862.1.1.1-5

Keywords : DNA-binding domain crystal structure, PSSM, SVM, MHC, epitope, peptide vaccine Abbreviations: Goldman, Engelberg and Steitz, (GES); major histocompatibility complex, (MHC); Position Specific Scoring Matrices, (PSSMs); Support Vector Machine, (SVM)
Conflict of Interest : None declared

The machine learning techniques are playing a major role in the field of immunoinformatics for DNA-binding domain analysis. Functional analysis of the binding ability of DNA-binding domain protein antigen peptides to major histocompatibility complex (MHC) class molecules is important in vaccine development. The variable length of each binding peptide complicates this prediction. Such predictions can be used to select epitopes for use in rational vaccine design and to increase the understanding of roles of the immune system in infectious diseases. Antigenic epitopes of DNA-binding domain protein form Human papilloma virus-31 are important determinant for protection of many host form viral infection. This study shows active part in host immune reactions and involvement of MHC class-I and MHC II in response to almost all antigens. We used PSSM and SVM algorithms for antigen design, which represented predicted binders as MHCII-IAb, MHCII-IAd, MHCII-IAg7, and MHCII- RT1.B nonamers from viral DNA-binding domain crystal structure. These peptide nonamers are from a set of aligned peptides known to bind to a given MHC molecule as the predictor of MHC-peptide binding. Analysis shows potential drug targets to identify active sites against diseases.

[1] Bhasin, M. and Raghava, G.P. (2005), “Pcleavage: an SVM based method for prediction of constitutive proteasome and immunoproteasome cleavage sites in antigenic sequences”. Nucleic Acids Research, 33, W202-207  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[2] Buus, S., et al., (2003), “Sensitive quantitative predictions of peptide- MHC binding by a 'Query by Committee' artificial neural network approach”. Tissue Antigens, 62, 378-384  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[3] Davidson, W., et al., (2004), “Characterization of the binding site for inhibitors of the HPV11 E1-E2 protein interaction on the E2 transactivation domain by photoaffinity labeling and mass spectrometry”. Anal Chem., 76(7), 2095-102.  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[4] Gomase, V.S. et al., (2006), “Automatic modeling of protein 3D structure Nucleoplasmin-like viral coat protein from Cucumber mosaic virus”. 14th International Conference on Advance Computing & Communication, Published by IEEE Computer Society in IEEE Xplore USA 614-615  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[5] Gomase, V.S. et al., (2007), "Prediction of MHC Binding Peptides and Epitopes from Alfalfa mosaic virus". Curr. Drug Discov. Technol., 4(2),117-1215, 2007  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[6] Gomase, V.S. et al., (2008a), "Prediction of MHC Binding Peptides and Epitopes from Groundnut Bud Necrosis Virus (GBNV)". J. of Proteomics & Bioinformatics, 1 (4), 188- 205.  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[7] Gomase, V.S. et al., (2008b), "Computer Aided Multi Parameter Antigen Design: Impact of Synthetic Peptide Vaccines from Soybean Mosaic Virus". ICETET 2008, IEEE Computer Society in IEEE Xplore, Los Alamitos, California, 629-634  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[8] Gomase, V.S. and Kale, K.V. (2008a), "In silico prediction of epitopes: a new approach for fragment based viral peptide vaccines". Int. J. of Applied Computing, 1(1), 39-46  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[9] Gomase, V.S. and Kale, K.V. (2008b), "Approach of proteomics system architecture in plant virus’s database". Int. J. of Applied Computing, 1(1), 33-38  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[10] Gomase, V.S. and Kale, K.V. (2008c), "Bioinformatics based sequence analysis of Nucleoplasmin like viral coat protein". Int. J. of Information Retrieval, 1(1), 11-15.  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[11] Nielsen, M., et al., (2003), “Reliable prediction of T-cell epitopes using neural networks with novel sequence representations”. Protein Sci., 12, 1007-1017  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[12] Sénéchal, H., et al., (2007), “Amino acid substitutions that specifically impair the transcriptional activity of papillomavirus E2 affect binding to the long isoform of Brd4”. Virology, 358(1), 10-7  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[13] Reche, P.A. , (2002), “Prediction of MHC Class I Binding Peptides Using Profile Motifs”. Hum Immun., 63, 701-709  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus  

[14] Wang, Y., et al., (2004), "Crystal structure of the E2 transactivation domain of human papillomavirus type 11 bound to a protein interaction inhibitor". J Biol Chem., 279(8), 6976-85  
» CrossRef   » Google Scholar   » PubMed   » DOAJ   » CAS   » Scopus