Ranking-based Convolutional Neural Network Models for Peptide-MHC Binding Prediction

TL;DR

This paper introduces ranking-based CNN models, ConvM and SpConvM, for peptide-MHC binding prediction, improving accuracy and robustness over existing methods by optimizing ranking objectives and incorporating position encoding.

Contribution

The authors develop novel ranking-based CNN models with position encoding for peptide-MHC binding prediction, outperforming state-of-the-art methods like MHCflurry.

Findings

Models significantly outperform MHCflurry with 6.70% higher AUC.

Achieved 17.10% improvement on ROC5 across 128 alleles.

Ranking-based optimization enhances robustness to measurement inaccuracies.

Abstract

T-cell receptors can recognize foreign peptides bound to major histocompatibility complex (MHC) class-I proteins, and thus trigger the adaptive immune response. Therefore, identifying peptides that can bind to MHC class-I molecules plays a vital role in the design of peptide vaccines. Many computational methods, for example, the state-of-the-art allele-specific method MHCflurry, have been developed to predict the binding affinities between peptides and MHC molecules. In this manuscript, we develop two allele-specific Convolutional Neural Network (CNN)-based methods named ConvM and SpConvM to tackle the binding prediction problem. Specifically, we formulate the problem as to optimize the rankings of peptide-MHC bindings via ranking-based learning objectives. Such optimization is more robust and tolerant to the measurement inaccuracy of binding affinities, and therefore enables more accurate prioritization of binding peptides. In addition, we develop a new position encoding method in ConvM and SpConvM to better identify the most important amino acids for the binding events. Our experimental results demonstrate that our models significantly outperform the state-of-the-art methods including MHCflurry with an average percentage improvement of 6.70% on AUC and 17.10% on ROC5 across 128 alleles.