Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Predicting Class II MHC-Peptide binding: a kernel based approach using similarity scores.

Jesper Salomon1, Darren R Flower

  • 1The Jenner Institute, University of Oxford, Compton, Newbury, Berkshire, RG20 7NN, UK. bio@salomons.dk

BMC Bioinformatics
|November 16, 2006
PubMed
Summary

This study introduces a novel kernel method for predicting peptide binding to Major Histocompatibility Complex (MHC) class II molecules. The approach effectively models variable-length peptides, improving prediction accuracy for T-cell epitope identification.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Supervised Learning of Protein Melting Temperature: Cross-Species vs. Species-Specific Prediction.

Proteins·2025
Same author

Expansion of the diversity of dispersin scaffolds.

Acta crystallographica. Section D, Structural biology·2025
Same author

A systematic analysis of regression models for protein engineering.

PLoS computational biology·2024
Same author

Extracting prime protein targets as possible drug candidates: machine learning evaluation.

Medical & biological engineering & computing·2023
Same author

Towards Effective Consensus Scoring in Structure-Based Virtual Screening.

Interdisciplinary sciences, computational life sciences·2022
Same author

To Affinity and Beyond: A Personal Reflection on the Design and Discovery of Drugs.

Molecules (Basel, Switzerland)·2022

Area of Science:

  • Immunoinformatics
  • Computational Biology
  • Molecular Modeling

Background:

  • Predicting peptide-MHC interactions is crucial for identifying T-cell epitopes.
  • Class II MHC alleles present challenges due to open binding grooves and variable peptide lengths.
  • Existing modeling methods struggle with the dynamic nature of peptide binding.

Purpose of the Study:

  • To develop a novel kernel method for accurate peptide binding prediction to MHC class II molecules.
  • To address the limitations of current methods in handling variable-length peptides.
  • To improve the identification of potential T-cell epitopes.

Main Methods:

  • A kernel method was developed to quantify similarities between peptide sequences.
  • Peptide similarities were integrated into the kernel to handle variable lengths.

Related Experiment Videos

  • The method was tested on established MHCPEP, MCHBN, and MHCBench datasets.

    • Main Results:

    • The kernel approach significantly improved prediction accuracy across multiple MHC class II alleles.
    • Higher numbers of true positives and negatives were observed compared to existing methods.
    • Cross-validation yielded an average AROC of 0.824 for MHCBench and 0.96 for MHCPEP datasets.

    Conclusions:

    • The novel method enhances performance over state-of-the-art MHC class II peptide binding predictions.
    • A custom, knowledge-based peptide representation using similarity scores offers flexibility.
    • This approach can be applied to other dynamic sequence-based modeling problems.