Neural network-based prediction of candidate T-cell epitopes

Margo C. Honeyman, Vladimir Brusic, Natalie L. Stone, Leonard C. Harrison

Research output: Contribution to journalArticle

155 Citations (Scopus)

Abstract

Activation of T cells requires recognition by T-cell receptors of specific peptides bound to major histocompatibility complex (MHC) molecules on the surface of either antigen-presenting or target cells. These peptides, T-cell epitopes, have potential therapeutic applications, such as for use as vaccines. Their identification, however, usually requires that multiple overlapping synthetic peptides encompassing a protein antigen be assayed, which in humans, is limited by volume of donor blood. T-cell epitopes are a subset of peptides that bind to MHC molecules. We use an artificial neural network (ANN) model trained to predict peptides that bind to the MHC class II molecule HLA-DR4(*0401). Binding prediction facilitates identification of T- cell epitopes in tyrosine phosphatase IA-2, an autoantigen in DR4-associated type 1 diabetes. Synthetic peptides encompassing IA-2 were tested experimentally for DR4 binding and T-cell proliferation in humans at risk for diabetes. ANN-based binding prediction was sensitive and specific, and reduced the number of peptides required for T-cell assay by more than half, with only a minor loss of epitopes. This strategy could expedite identification of candidate T-cell epitopes in diverse diseases.

Original languageEnglish
Pages (from-to)966-969
Number of pages4
JournalNature Biotechnology
Volume16
Issue number10
DOIs
Publication statusPublished - 1998
Externally publishedYes

Fingerprint

Epitopes
T-Lymphocyte Epitopes
T-cells
Peptides
Neural networks
Major Histocompatibility Complex
Peptide T
T-Lymphocytes
Medical problems
Class 8 Receptor-Like Protein Tyrosine Phosphatases
Molecules
HLA-DR4 Antigen
Antigens
Neural Networks (Computer)
Autoantigens
Surface Antigens
T-Cell Antigen Receptor
Blood Volume
Blood Donors
Type 1 Diabetes Mellitus

Keywords

  • Applied immunology
  • Bioinformatics
  • MHC

ASJC Scopus subject areas

  • Microbiology

Cite this

Honeyman, M. C., Brusic, V., Stone, N. L., & Harrison, L. C. (1998). Neural network-based prediction of candidate T-cell epitopes. Nature Biotechnology, 16(10), 966-969. https://doi.org/10.1038/nbt1098-966

Neural network-based prediction of candidate T-cell epitopes. / Honeyman, Margo C.; Brusic, Vladimir; Stone, Natalie L.; Harrison, Leonard C.

In: Nature Biotechnology, Vol. 16, No. 10, 1998, p. 966-969.

Research output: Contribution to journalArticle

Honeyman, MC, Brusic, V, Stone, NL & Harrison, LC 1998, 'Neural network-based prediction of candidate T-cell epitopes', Nature Biotechnology, vol. 16, no. 10, pp. 966-969. https://doi.org/10.1038/nbt1098-966
Honeyman, Margo C. ; Brusic, Vladimir ; Stone, Natalie L. ; Harrison, Leonard C. / Neural network-based prediction of candidate T-cell epitopes. In: Nature Biotechnology. 1998 ; Vol. 16, No. 10. pp. 966-969.
@article{9a9ba8f258e8404f9ef02a808f79955e,
title = "Neural network-based prediction of candidate T-cell epitopes",
abstract = "Activation of T cells requires recognition by T-cell receptors of specific peptides bound to major histocompatibility complex (MHC) molecules on the surface of either antigen-presenting or target cells. These peptides, T-cell epitopes, have potential therapeutic applications, such as for use as vaccines. Their identification, however, usually requires that multiple overlapping synthetic peptides encompassing a protein antigen be assayed, which in humans, is limited by volume of donor blood. T-cell epitopes are a subset of peptides that bind to MHC molecules. We use an artificial neural network (ANN) model trained to predict peptides that bind to the MHC class II molecule HLA-DR4(*0401). Binding prediction facilitates identification of T- cell epitopes in tyrosine phosphatase IA-2, an autoantigen in DR4-associated type 1 diabetes. Synthetic peptides encompassing IA-2 were tested experimentally for DR4 binding and T-cell proliferation in humans at risk for diabetes. ANN-based binding prediction was sensitive and specific, and reduced the number of peptides required for T-cell assay by more than half, with only a minor loss of epitopes. This strategy could expedite identification of candidate T-cell epitopes in diverse diseases.",
keywords = "Applied immunology, Bioinformatics, MHC",
author = "Honeyman, {Margo C.} and Vladimir Brusic and Stone, {Natalie L.} and Harrison, {Leonard C.}",
year = "1998",
doi = "10.1038/nbt1098-966",
language = "English",
volume = "16",
pages = "966--969",
journal = "Nature Biotechnology",
issn = "1087-0156",
publisher = "Nature Publishing Group",
number = "10",

}

TY - JOUR

T1 - Neural network-based prediction of candidate T-cell epitopes

AU - Honeyman, Margo C.

AU - Brusic, Vladimir

AU - Stone, Natalie L.

AU - Harrison, Leonard C.

PY - 1998

Y1 - 1998

N2 - Activation of T cells requires recognition by T-cell receptors of specific peptides bound to major histocompatibility complex (MHC) molecules on the surface of either antigen-presenting or target cells. These peptides, T-cell epitopes, have potential therapeutic applications, such as for use as vaccines. Their identification, however, usually requires that multiple overlapping synthetic peptides encompassing a protein antigen be assayed, which in humans, is limited by volume of donor blood. T-cell epitopes are a subset of peptides that bind to MHC molecules. We use an artificial neural network (ANN) model trained to predict peptides that bind to the MHC class II molecule HLA-DR4(*0401). Binding prediction facilitates identification of T- cell epitopes in tyrosine phosphatase IA-2, an autoantigen in DR4-associated type 1 diabetes. Synthetic peptides encompassing IA-2 were tested experimentally for DR4 binding and T-cell proliferation in humans at risk for diabetes. ANN-based binding prediction was sensitive and specific, and reduced the number of peptides required for T-cell assay by more than half, with only a minor loss of epitopes. This strategy could expedite identification of candidate T-cell epitopes in diverse diseases.

AB - Activation of T cells requires recognition by T-cell receptors of specific peptides bound to major histocompatibility complex (MHC) molecules on the surface of either antigen-presenting or target cells. These peptides, T-cell epitopes, have potential therapeutic applications, such as for use as vaccines. Their identification, however, usually requires that multiple overlapping synthetic peptides encompassing a protein antigen be assayed, which in humans, is limited by volume of donor blood. T-cell epitopes are a subset of peptides that bind to MHC molecules. We use an artificial neural network (ANN) model trained to predict peptides that bind to the MHC class II molecule HLA-DR4(*0401). Binding prediction facilitates identification of T- cell epitopes in tyrosine phosphatase IA-2, an autoantigen in DR4-associated type 1 diabetes. Synthetic peptides encompassing IA-2 were tested experimentally for DR4 binding and T-cell proliferation in humans at risk for diabetes. ANN-based binding prediction was sensitive and specific, and reduced the number of peptides required for T-cell assay by more than half, with only a minor loss of epitopes. This strategy could expedite identification of candidate T-cell epitopes in diverse diseases.

KW - Applied immunology

KW - Bioinformatics

KW - MHC

UR - http://www.scopus.com/inward/record.url?scp=0031760687&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031760687&partnerID=8YFLogxK

U2 - 10.1038/nbt1098-966

DO - 10.1038/nbt1098-966

M3 - Article

C2 - 9788355

AN - SCOPUS:0031760687

VL - 16

SP - 966

EP - 969

JO - Nature Biotechnology

JF - Nature Biotechnology

SN - 1087-0156

IS - 10

ER -