Limitations of Ab Initio Predictions of Peptide Binding to MHC Class II Molecules

Hao Zhang,Ying Xu,Alessandro Sette,Peng Wang,Ole Lund,Julia Ponomarenko,Morten Nielsen,Bjoern Peters,Philip E Bourne,Nikitas Papangelopoulos,Narcis Fernandez-Fuentes

doi:10.1371/journal.pone.0009272

Abstract

Successful predictions of peptide MHC binding typically require a large set of binding data for the specific MHC molecule that is examined. Structure based prediction methods promise to circumvent this requirement by evaluating the physical contacts a peptide can make with an MHC molecule based on the highly conserved 3D structure of peptide:MHC complexes. While several such methods have been described before, most are not publicly available and have not been independently tested for their performance. We here implemented and evaluated three prediction methods for MHC class II molecules: statistical potentials derived from the analysis of known protein structures; energetic evaluation of different peptide snapshots in a molecular dynamics simulation; and direct analysis of contacts made in known 3D structures of peptide:MHC complexes. These methods are ab initio in that they require structural data of the MHC molecule examined, but no specific peptide:MHC binding data. Moreover, these methods retain the ability to make predictions in a sufficiently short time scale to be useful in a real world application, such as screening a whole proteome for candidate binding peptides. A rigorous evaluation of each methods prediction performance showed that these are significantly better than random, but still substantially lower than the best performing sequence based class II prediction methods available. While the approaches presented here were developed independently, we have chosen to present our results together in order to support the notion that generating structure based predictions of peptide:MHC binding without using binding data is unlikely to give satisfactory results.

Highlights

A common bioinformatics application in immunology is the prediction of peptide binding to MHC molecules [1]
The approaches are based on 1) statistical potentials derived from the analysis of known protein structures, 2) energetic evaluation of different peptide snapshots in a molecular dynamics simulation, and 3) direct analysis of contacts made in known 3D structures of peptide:MHC complexes
This section is separated into two parts: In the first part, results are reported that were generated during the derivation of each of the three structure-based prediction methods, starting with the statistical pair potential-based method, followed by the molecular dynamics simulation and the contact map-based method

Summary

Introduction

A common bioinformatics application in immunology is the prediction of peptide binding to MHC molecules [1] Most such binding predictions are based on machine learning algorithms, which aim to generalize experimental binding data to define a binding sequence pattern for a given MHC molecule. A structure-based predictive understanding of peptide:MHC binding provides a physical explanation for the nature of the binding interactions, while purely peptide sequence based learning methods merely provide a description of the sequence characteristics of preferred MHC-binding ligands. Throughout this manuscript, we refer to prediction approaches that use structural information but not peptide:MHC binding data as ‘ab initio’ approaches

Methods

Results

Conclusion