Methods for prediction of peptide binding to MHC molecules: A comparative study

Kun Yu, Nikolai Petrovsky, Christian Schönbach, Judice L Y Koh, Vladimir Brusic

Research output: Contribution to journalArticle

114 Citations (Scopus)

Abstract

Background: A variety of methods for prediction of peptide binding to major histocompatibility complex (MHC) have been proposed. These methods are based on binding motifs, binding matrices, hidden Markov models (HMM), or artificial neural networks (ANN). There has been little prior work on the comparative analysis of these methods. Materials and Methods: We performed a comparison of the performance of six methods applied to the prediction of two human MHC class I molecules, including binding matrices and motifs, ANNs, and HMMs. Results: The selection of the optimal prediction method depends on the amount of available data (the number of peptides of known binding affinity to the MHC molecule of interest), the biases in the data set and the intended purpose of the prediction (screening of a single protein versus mass screening). When little or no peptide data are available, binding motifs are the most useful alternative to random guessing or use of a complete overlapping set of peptides for selection of candidate binders. As the number of known peptide binders increases, binding matrices and HMM become more useful predictors. ANN and HMM are the predictive methods of choice for MHC alleles with more than 100 known binding peptides. Conclusion: The ability of bioinformatic methods to reliably predict MHC binding peptides, and thereby potential T-cell epitopes, has major implications for clinical immunology, particularly in the area of vaccine design.

Original languageEnglish
Pages (from-to)137-148
Number of pages12
JournalMolecular Medicine
Volume8
Issue number3
Publication statusPublished - 2002
Externally publishedYes

Fingerprint

Major Histocompatibility Complex
Peptides
Mass Screening
T-Lymphocyte Epitopes
Allergy and Immunology
Computational Biology
Vaccines
Alleles
Proteins

ASJC Scopus subject areas

  • Genetics

Cite this

Yu, K., Petrovsky, N., Schönbach, C., Koh, J. L. Y., & Brusic, V. (2002). Methods for prediction of peptide binding to MHC molecules: A comparative study. Molecular Medicine, 8(3), 137-148.

Methods for prediction of peptide binding to MHC molecules : A comparative study. / Yu, Kun; Petrovsky, Nikolai; Schönbach, Christian; Koh, Judice L Y; Brusic, Vladimir.

In: Molecular Medicine, Vol. 8, No. 3, 2002, p. 137-148.

Research output: Contribution to journalArticle

Yu, K, Petrovsky, N, Schönbach, C, Koh, JLY & Brusic, V 2002, 'Methods for prediction of peptide binding to MHC molecules: A comparative study', Molecular Medicine, vol. 8, no. 3, pp. 137-148.
Yu, Kun ; Petrovsky, Nikolai ; Schönbach, Christian ; Koh, Judice L Y ; Brusic, Vladimir. / Methods for prediction of peptide binding to MHC molecules : A comparative study. In: Molecular Medicine. 2002 ; Vol. 8, No. 3. pp. 137-148.
@article{601bbe2b8bd24da187bab768bb7c31c3,
title = "Methods for prediction of peptide binding to MHC molecules: A comparative study",
abstract = "Background: A variety of methods for prediction of peptide binding to major histocompatibility complex (MHC) have been proposed. These methods are based on binding motifs, binding matrices, hidden Markov models (HMM), or artificial neural networks (ANN). There has been little prior work on the comparative analysis of these methods. Materials and Methods: We performed a comparison of the performance of six methods applied to the prediction of two human MHC class I molecules, including binding matrices and motifs, ANNs, and HMMs. Results: The selection of the optimal prediction method depends on the amount of available data (the number of peptides of known binding affinity to the MHC molecule of interest), the biases in the data set and the intended purpose of the prediction (screening of a single protein versus mass screening). When little or no peptide data are available, binding motifs are the most useful alternative to random guessing or use of a complete overlapping set of peptides for selection of candidate binders. As the number of known peptide binders increases, binding matrices and HMM become more useful predictors. ANN and HMM are the predictive methods of choice for MHC alleles with more than 100 known binding peptides. Conclusion: The ability of bioinformatic methods to reliably predict MHC binding peptides, and thereby potential T-cell epitopes, has major implications for clinical immunology, particularly in the area of vaccine design.",
author = "Kun Yu and Nikolai Petrovsky and Christian Sch{\"o}nbach and Koh, {Judice L Y} and Vladimir Brusic",
year = "2002",
language = "English",
volume = "8",
pages = "137--148",
journal = "Molecular Medicine",
issn = "1076-1551",
publisher = "Feinstein Institute for Medical Research",
number = "3",

}

TY - JOUR

T1 - Methods for prediction of peptide binding to MHC molecules

T2 - A comparative study

AU - Yu, Kun

AU - Petrovsky, Nikolai

AU - Schönbach, Christian

AU - Koh, Judice L Y

AU - Brusic, Vladimir

PY - 2002

Y1 - 2002

N2 - Background: A variety of methods for prediction of peptide binding to major histocompatibility complex (MHC) have been proposed. These methods are based on binding motifs, binding matrices, hidden Markov models (HMM), or artificial neural networks (ANN). There has been little prior work on the comparative analysis of these methods. Materials and Methods: We performed a comparison of the performance of six methods applied to the prediction of two human MHC class I molecules, including binding matrices and motifs, ANNs, and HMMs. Results: The selection of the optimal prediction method depends on the amount of available data (the number of peptides of known binding affinity to the MHC molecule of interest), the biases in the data set and the intended purpose of the prediction (screening of a single protein versus mass screening). When little or no peptide data are available, binding motifs are the most useful alternative to random guessing or use of a complete overlapping set of peptides for selection of candidate binders. As the number of known peptide binders increases, binding matrices and HMM become more useful predictors. ANN and HMM are the predictive methods of choice for MHC alleles with more than 100 known binding peptides. Conclusion: The ability of bioinformatic methods to reliably predict MHC binding peptides, and thereby potential T-cell epitopes, has major implications for clinical immunology, particularly in the area of vaccine design.

AB - Background: A variety of methods for prediction of peptide binding to major histocompatibility complex (MHC) have been proposed. These methods are based on binding motifs, binding matrices, hidden Markov models (HMM), or artificial neural networks (ANN). There has been little prior work on the comparative analysis of these methods. Materials and Methods: We performed a comparison of the performance of six methods applied to the prediction of two human MHC class I molecules, including binding matrices and motifs, ANNs, and HMMs. Results: The selection of the optimal prediction method depends on the amount of available data (the number of peptides of known binding affinity to the MHC molecule of interest), the biases in the data set and the intended purpose of the prediction (screening of a single protein versus mass screening). When little or no peptide data are available, binding motifs are the most useful alternative to random guessing or use of a complete overlapping set of peptides for selection of candidate binders. As the number of known peptide binders increases, binding matrices and HMM become more useful predictors. ANN and HMM are the predictive methods of choice for MHC alleles with more than 100 known binding peptides. Conclusion: The ability of bioinformatic methods to reliably predict MHC binding peptides, and thereby potential T-cell epitopes, has major implications for clinical immunology, particularly in the area of vaccine design.

UR - http://www.scopus.com/inward/record.url?scp=0036042580&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036042580&partnerID=8YFLogxK

M3 - Article

C2 - 12142545

AN - SCOPUS:0036042580

VL - 8

SP - 137

EP - 148

JO - Molecular Medicine

JF - Molecular Medicine

SN - 1076-1551

IS - 3

ER -