Hidden Markov model-based prediction of antigenic peptides that interact with MHC class II molecules

Hideki Noguchi, Ryuji Kato, Taizo Hanai, Yukari Matsubara, Hiroyuki Honda, Vladimir Brusic, Takeshi Kobayashi

Research output: Contribution to journalArticle

66 Citations (Scopus)

Abstract

Elucidating the interaction between major histocompatibility complex (MHC) molecules and antigenic peptides is fundamental to better understanding of the processes involved in immune responses and for the development of innovative immunotherapies. In the present study, hidden Markov models (HMM) were combined with the successive state splitting (SSS) algorithm for optimization of the HMM structure, to predict peptide binders to the human MHC class II molecule HLA-DRBI*0101. The predictive performance of our model (S-HMM) was compared with fully connected HMM and artificial neural network (ANN) methods using the relative operating characteristic (ROC) analysis. The S-HMM predictions had values of ROC≥0.85 which was at least as good, or better than the comparison methods. In addition, S-HMM is trained on positive data only and does not require exhaustive data preprocessing, such as peptide alignment. Our results demonstrated that S-HMM combines the high accuracy of predictions with the simplicity of implementation and is therefore useful for analyzing MHC class II binding peptides. In particular the S-HMM may be trained using only positive data and, the preprocessing of training data, such as peptide alignment and the selection of binding cores, is not required in this method.

Original languageEnglish
Pages (from-to)264-270
Number of pages7
JournalJournal of Bioscience and Bioengineering
Volume94
Issue number3
DOIs
Publication statusPublished - 2002
Externally publishedYes

Fingerprint

Hidden Markov models
Major Histocompatibility Complex
Peptides
Molecules
Neural Networks (Computer)
Immunotherapy
Model structures
Binders
Neural networks

Keywords

  • Binding peptides
  • Bioinformatics
  • Hidden Markov model
  • MHC class II
  • Successive state splitting

ASJC Scopus subject areas

  • Biotechnology
  • Bioengineering

Cite this

Hidden Markov model-based prediction of antigenic peptides that interact with MHC class II molecules. / Noguchi, Hideki; Kato, Ryuji; Hanai, Taizo; Matsubara, Yukari; Honda, Hiroyuki; Brusic, Vladimir; Kobayashi, Takeshi.

In: Journal of Bioscience and Bioengineering, Vol. 94, No. 3, 2002, p. 264-270.

Research output: Contribution to journalArticle

Noguchi, Hideki ; Kato, Ryuji ; Hanai, Taizo ; Matsubara, Yukari ; Honda, Hiroyuki ; Brusic, Vladimir ; Kobayashi, Takeshi. / Hidden Markov model-based prediction of antigenic peptides that interact with MHC class II molecules. In: Journal of Bioscience and Bioengineering. 2002 ; Vol. 94, No. 3. pp. 264-270.
@article{1b4bcd23599342e3b856a0b06a252d36,
title = "Hidden Markov model-based prediction of antigenic peptides that interact with MHC class II molecules",
abstract = "Elucidating the interaction between major histocompatibility complex (MHC) molecules and antigenic peptides is fundamental to better understanding of the processes involved in immune responses and for the development of innovative immunotherapies. In the present study, hidden Markov models (HMM) were combined with the successive state splitting (SSS) algorithm for optimization of the HMM structure, to predict peptide binders to the human MHC class II molecule HLA-DRBI*0101. The predictive performance of our model (S-HMM) was compared with fully connected HMM and artificial neural network (ANN) methods using the relative operating characteristic (ROC) analysis. The S-HMM predictions had values of ROC≥0.85 which was at least as good, or better than the comparison methods. In addition, S-HMM is trained on positive data only and does not require exhaustive data preprocessing, such as peptide alignment. Our results demonstrated that S-HMM combines the high accuracy of predictions with the simplicity of implementation and is therefore useful for analyzing MHC class II binding peptides. In particular the S-HMM may be trained using only positive data and, the preprocessing of training data, such as peptide alignment and the selection of binding cores, is not required in this method.",
keywords = "Binding peptides, Bioinformatics, Hidden Markov model, MHC class II, Successive state splitting",
author = "Hideki Noguchi and Ryuji Kato and Taizo Hanai and Yukari Matsubara and Hiroyuki Honda and Vladimir Brusic and Takeshi Kobayashi",
year = "2002",
doi = "10.1016/S1389-1723(02)80160-8",
language = "English",
volume = "94",
pages = "264--270",
journal = "Journal of Bioscience and Bioengineering",
issn = "1389-1723",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Hidden Markov model-based prediction of antigenic peptides that interact with MHC class II molecules

AU - Noguchi, Hideki

AU - Kato, Ryuji

AU - Hanai, Taizo

AU - Matsubara, Yukari

AU - Honda, Hiroyuki

AU - Brusic, Vladimir

AU - Kobayashi, Takeshi

PY - 2002

Y1 - 2002

N2 - Elucidating the interaction between major histocompatibility complex (MHC) molecules and antigenic peptides is fundamental to better understanding of the processes involved in immune responses and for the development of innovative immunotherapies. In the present study, hidden Markov models (HMM) were combined with the successive state splitting (SSS) algorithm for optimization of the HMM structure, to predict peptide binders to the human MHC class II molecule HLA-DRBI*0101. The predictive performance of our model (S-HMM) was compared with fully connected HMM and artificial neural network (ANN) methods using the relative operating characteristic (ROC) analysis. The S-HMM predictions had values of ROC≥0.85 which was at least as good, or better than the comparison methods. In addition, S-HMM is trained on positive data only and does not require exhaustive data preprocessing, such as peptide alignment. Our results demonstrated that S-HMM combines the high accuracy of predictions with the simplicity of implementation and is therefore useful for analyzing MHC class II binding peptides. In particular the S-HMM may be trained using only positive data and, the preprocessing of training data, such as peptide alignment and the selection of binding cores, is not required in this method.

AB - Elucidating the interaction between major histocompatibility complex (MHC) molecules and antigenic peptides is fundamental to better understanding of the processes involved in immune responses and for the development of innovative immunotherapies. In the present study, hidden Markov models (HMM) were combined with the successive state splitting (SSS) algorithm for optimization of the HMM structure, to predict peptide binders to the human MHC class II molecule HLA-DRBI*0101. The predictive performance of our model (S-HMM) was compared with fully connected HMM and artificial neural network (ANN) methods using the relative operating characteristic (ROC) analysis. The S-HMM predictions had values of ROC≥0.85 which was at least as good, or better than the comparison methods. In addition, S-HMM is trained on positive data only and does not require exhaustive data preprocessing, such as peptide alignment. Our results demonstrated that S-HMM combines the high accuracy of predictions with the simplicity of implementation and is therefore useful for analyzing MHC class II binding peptides. In particular the S-HMM may be trained using only positive data and, the preprocessing of training data, such as peptide alignment and the selection of binding cores, is not required in this method.

KW - Binding peptides

KW - Bioinformatics

KW - Hidden Markov model

KW - MHC class II

KW - Successive state splitting

UR - http://www.scopus.com/inward/record.url?scp=0036399733&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036399733&partnerID=8YFLogxK

U2 - 10.1016/S1389-1723(02)80160-8

DO - 10.1016/S1389-1723(02)80160-8

M3 - Article

VL - 94

SP - 264

EP - 270

JO - Journal of Bioscience and Bioengineering

JF - Journal of Bioscience and Bioengineering

SN - 1389-1723

IS - 3

ER -