Prediction Of Antigenic Conformational Epitope Patches And Residues On The Protein 3d Structures  

Abstract Category: Science
Course / Degree: MASTER of Science in Computer Science
Institution / University: AASTMT, Egypt
Published in: 2012

Dissertation Abstract / Summary:

Vaccination is a method of stimulating resistant in the human body to specific diseases using microorganisms (bacteria or viruses) that have been modified or killed. Currently, the available vaccines are composed of killed or life attenuated the whole pathogens before being injected to the human body. Despite the advantages of such vaccines, there are a lot of potential safety problems such as incomplete inactivation of the killed pathogens or the evolution of the attenuated virus. Instead of the entire microbe, subunit vaccines include only the antigens that best stimulate the immune system. This type of vaccine uses epitopes – the very specific parts of the antigen that antibodies or T cells recognize and bind. Identification of epitopes that invoke strong responses from B-cells is one of the key steps in designing effective vaccines against pathogens. Because of the experimental determination of epitopes is expensive in terms of cost, time, and effort involved, there is an urgent need for computational methods to reliably identify the B-cell epitopes.

In this thesis, we developed a novel computational method “CBCPRED” to predict the conformational B-cell epitope residues from the target antigen protein structure. The method combines the support vector machine model with the protein structural features and the propensity score of amino acid physico-chemical properties. Using fivefold and leave-one-out cross validation techniques on the 75 antigen structures of the DiscoTope dataset; CBCPRED achieved an area under receiver operator characteristics curve (AUC) of 0.818 and 0.859, respectively. Using independent testing on benchmark dataset after removing antigens sequence redundancy, CBCPRED achieved AUC of 0.747. Comparison with the existing methods shows that the prediction performance of CBCPRED to identify the conformation B-cell epitope residues from the antigen 3D structure is higher compared to the best results of existing methods by about 0.02 in case of fivefold cross validation, 0.041 in case of leave-one-out cross validation, and 0.01 in case of independent testing. Experimental results indicate that besides the well-known structural features, the amino acid physico-chemical information can also be used to identify conformational epitopes. Furthermore, combination of these features can remarkably improve the prediction performance. CBCPRED is now available online.

In addition, we proposed – PatchTope – A different vision in the identification of the antigenic epitope sites in the antigen structure chain by predicting the non-overlapping surface patches that hold the majority of epitope residues in the target antigen structure and hence, the scientists can use the predicted patches in vaccine development. Predictions are made for the known structures of benchmark dataset after removing antigens sequence redundancy. The predictions were successful for 70% of the antigen structure chains of the benchmark dataset. We compared the prediction performance of PatchTope model with a protein-protein interaction prediction server “Sharp2” which predicts the protein interacting sites on the surface of the protein structure, and using the same antigen structures of the benchmark dataset, it was observed that our model outperforms on Sharp2 by more than 40% accuracy. A web server “PatchTope” has been developed for predicting antigenic epitope surface patches on an antigen protein structure surface and is available online.

Dissertation Keywords/Search Tags:
Conformational B-cell epitope, Discontinuous B-Cell epitope, Epitope Prediction, Immunology, Immunoinformatics, Residues, Patch, Protein structure, PDB, CBCPRED, PatchTope

This Dissertation Abstract may be cited as follows:
No user preference. Please use the standard reference methodology.

Submission Details: Dissertation Abstract submitted by Khaled Hassan from Egypt on 01-May-2012 19:57.
Abstract has been viewed 2486 times (since 7 Mar 2010).

Khaled Hassan Contact Details: Email: khaled.hassan.edu@gmail.com Phone: +2 0111 2173377

Great care has been taken to ensure that this information is correct, however ThesisAbstracts.com cannot accept responsibility for the contents of this Dissertation abstract titled "Prediction Of Antigenic Conformational Epitope Patches And Residues On The Protein 3d Structures". This abstract has been submitted by Khaled Hassan on 01-May-2012 19:57. You may report a problem using the contact form.
© Copyright 2003 - 2023 of ThesisAbstracts.com and respective owners.

Copyright © Thesis Abstract | Dissertation Abstracts Thesis Library 2003-2023.
by scope.com.mt @ website design