Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine

Guang Lan Zhang, Joo Chuan Tong, Zong Hong Zhang, Yun Zheng, J. Thomas August, Chee Keong Kwoh, Vladimir Brusic

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Computational vaccinology is a developing discipline. To become a standard component in vaccine development, it requires accurate and broadly applicable models of wet-lab experiments. We developed prediction models based on a novel data representation of peptide/MHC interaction and support vector machines (SVM) for prediction of peptides that promiscuously bind to multiple Human Leukocyte Antigen (HLA) alleles belonging to HLA-B7 supertype. 10-fold cross-validation results showed that the area under the receiver operating curve (Aroc) of SVM models is above 0.90. Blind testing results showed that the average Aroc of SVM models is 0.84. A learning approach based on information theory, termed Information Learning Approach, was used for feature selection. Several amino acid positions with high information content have been identified in input 9mer peptides and HLA alleles and were used as input features to SVM. They are position 1, 2, 4, 5, 7, 8, 9 in 9mer peptides and position 45 and 97 in HLA-B7 molecules. Prediction accuracy was improved after feature selection. These positions cover the anchor positions of HLA-B7 alleles, which have important biological roles for successful biding of relevant peptides.

Original languageEnglish
Title of host publicationICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering
Pages319-323
Number of pages5
DOIs
Publication statusPublished - 2006
Externally publishedYes
EventICBPE 2006 - 2006 International Conference on Biomedical and Pharmaceutical Engineering - Singapore, Singapore
Duration: Dec 11 2006Dec 14 2006

Other

OtherICBPE 2006 - 2006 International Conference on Biomedical and Pharmaceutical Engineering
CountrySingapore
CitySingapore
Period12/11/0612/14/06

Fingerprint

Information Theory
Information theory
Antigens
HLA Antigens
Peptides
Binders
Support vector machines
Alleles
Feature extraction
Learning
B7 Antigens
Vaccines
Anchors
Amino acids
Support Vector Machine
Amino Acids
Molecules
Testing
Experiments

Keywords

  • Binding peptide
  • HLA-B7
  • Information thoery
  • Support vector machine
  • Vaccinology

ASJC Scopus subject areas

  • Biomedical Engineering
  • Pharmacology (medical)
  • Pharmacology, Toxicology and Pharmaceutics(all)

Cite this

Zhang, G. L., Tong, J. C., Zhang, Z. H., Zheng, Y., August, J. T., Kwoh, C. K., & Brusic, V. (2006). Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine. In ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering (pp. 319-323). [4155916] https://doi.org/10.1109/ICBPE.2006.348607

Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine. / Zhang, Guang Lan; Tong, Joo Chuan; Zhang, Zong Hong; Zheng, Yun; August, J. Thomas; Kwoh, Chee Keong; Brusic, Vladimir.

ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering. 2006. p. 319-323 4155916.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, GL, Tong, JC, Zhang, ZH, Zheng, Y, August, JT, Kwoh, CK & Brusic, V 2006, Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine. in ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering., 4155916, pp. 319-323, ICBPE 2006 - 2006 International Conference on Biomedical and Pharmaceutical Engineering, Singapore, Singapore, 12/11/06. https://doi.org/10.1109/ICBPE.2006.348607
Zhang GL, Tong JC, Zhang ZH, Zheng Y, August JT, Kwoh CK et al. Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine. In ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering. 2006. p. 319-323. 4155916 https://doi.org/10.1109/ICBPE.2006.348607
Zhang, Guang Lan ; Tong, Joo Chuan ; Zhang, Zong Hong ; Zheng, Yun ; August, J. Thomas ; Kwoh, Chee Keong ; Brusic, Vladimir. / Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine. ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering. 2006. pp. 319-323
@inproceedings{731b0ac3b56a4e8d8b3e8933474f7ea7,
title = "Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine",
abstract = "Computational vaccinology is a developing discipline. To become a standard component in vaccine development, it requires accurate and broadly applicable models of wet-lab experiments. We developed prediction models based on a novel data representation of peptide/MHC interaction and support vector machines (SVM) for prediction of peptides that promiscuously bind to multiple Human Leukocyte Antigen (HLA) alleles belonging to HLA-B7 supertype. 10-fold cross-validation results showed that the area under the receiver operating curve (Aroc) of SVM models is above 0.90. Blind testing results showed that the average Aroc of SVM models is 0.84. A learning approach based on information theory, termed Information Learning Approach, was used for feature selection. Several amino acid positions with high information content have been identified in input 9mer peptides and HLA alleles and were used as input features to SVM. They are position 1, 2, 4, 5, 7, 8, 9 in 9mer peptides and position 45 and 97 in HLA-B7 molecules. Prediction accuracy was improved after feature selection. These positions cover the anchor positions of HLA-B7 alleles, which have important biological roles for successful biding of relevant peptides.",
keywords = "Binding peptide, HLA-B7, Information thoery, Support vector machine, Vaccinology",
author = "Zhang, {Guang Lan} and Tong, {Joo Chuan} and Zhang, {Zong Hong} and Yun Zheng and August, {J. Thomas} and Kwoh, {Chee Keong} and Vladimir Brusic",
year = "2006",
doi = "10.1109/ICBPE.2006.348607",
language = "English",
isbn = "8190426249",
pages = "319--323",
booktitle = "ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering",

}

TY - GEN

T1 - Computational models for identifying promiscuous HLA-B7 binders based on information theory and support vector machine

AU - Zhang, Guang Lan

AU - Tong, Joo Chuan

AU - Zhang, Zong Hong

AU - Zheng, Yun

AU - August, J. Thomas

AU - Kwoh, Chee Keong

AU - Brusic, Vladimir

PY - 2006

Y1 - 2006

N2 - Computational vaccinology is a developing discipline. To become a standard component in vaccine development, it requires accurate and broadly applicable models of wet-lab experiments. We developed prediction models based on a novel data representation of peptide/MHC interaction and support vector machines (SVM) for prediction of peptides that promiscuously bind to multiple Human Leukocyte Antigen (HLA) alleles belonging to HLA-B7 supertype. 10-fold cross-validation results showed that the area under the receiver operating curve (Aroc) of SVM models is above 0.90. Blind testing results showed that the average Aroc of SVM models is 0.84. A learning approach based on information theory, termed Information Learning Approach, was used for feature selection. Several amino acid positions with high information content have been identified in input 9mer peptides and HLA alleles and were used as input features to SVM. They are position 1, 2, 4, 5, 7, 8, 9 in 9mer peptides and position 45 and 97 in HLA-B7 molecules. Prediction accuracy was improved after feature selection. These positions cover the anchor positions of HLA-B7 alleles, which have important biological roles for successful biding of relevant peptides.

AB - Computational vaccinology is a developing discipline. To become a standard component in vaccine development, it requires accurate and broadly applicable models of wet-lab experiments. We developed prediction models based on a novel data representation of peptide/MHC interaction and support vector machines (SVM) for prediction of peptides that promiscuously bind to multiple Human Leukocyte Antigen (HLA) alleles belonging to HLA-B7 supertype. 10-fold cross-validation results showed that the area under the receiver operating curve (Aroc) of SVM models is above 0.90. Blind testing results showed that the average Aroc of SVM models is 0.84. A learning approach based on information theory, termed Information Learning Approach, was used for feature selection. Several amino acid positions with high information content have been identified in input 9mer peptides and HLA alleles and were used as input features to SVM. They are position 1, 2, 4, 5, 7, 8, 9 in 9mer peptides and position 45 and 97 in HLA-B7 molecules. Prediction accuracy was improved after feature selection. These positions cover the anchor positions of HLA-B7 alleles, which have important biological roles for successful biding of relevant peptides.

KW - Binding peptide

KW - HLA-B7

KW - Information thoery

KW - Support vector machine

KW - Vaccinology

UR - http://www.scopus.com/inward/record.url?scp=46249107122&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46249107122&partnerID=8YFLogxK

U2 - 10.1109/ICBPE.2006.348607

DO - 10.1109/ICBPE.2006.348607

M3 - Conference contribution

SN - 8190426249

SN - 9788190426244

SP - 319

EP - 323

BT - ICBPE 2006 - Proceedings of the 2006 International Conference on Biomedical and Pharmaceutical Engineering

ER -