Improvement of Neoantigen Identification Through Convolution Neural Network.

Qing Hao,Heng Xu,Ping Wei,Yang Shu,Jun-Ning Zhao,Yi-Guan Zhang

doi:10.3389/fimmu.2021.682103

Qing Hao, Heng Xu + Show 4 more

Open Access

https://doi.org/10.3389/fimmu.2021.682103

Copy DOI

Abstract

Accurate prediction of neoantigens and the subsequent elicited protective anti-tumor response are particularly important for the development of cancer vaccine and adoptive T-cell therapy. However, current algorithms for predicting neoantigens are limited by in vitro binding affinity data and algorithmic constraints, inevitably resulting in high false positives. In this study, we proposed a deep convolutional neural network named APPM (antigen presentation prediction model) to predict antigen presentation in the context of human leukocyte antigen (HLA) class I alleles. APPM is trained on large mass spectrometry (MS) HLA-peptides datasets and evaluated with an independent MS benchmark. Results show that APPM outperforms the methods recommended by the immune epitope database (IEDB) in terms of positive predictive value (PPV) (0.40 vs. 0.22), which will further increase after combining these two approaches (PPV = 0.51). We further applied our model to the prediction of neoantigens from consensus driver mutations and identified 16,000 putative neoantigens with hallmarks of ‘drivers’.

Highlights

Cancer develops as a result of the accumulation of tumor-specific somatic mutations [1,2,3], where non-silent mutations in the coding region could be recognized as beacons of “foreign” by the immune system, named neoantigen [4, 5]
We proposed an antigen presentation prediction model (APPM), a Convolutional Neural Network (CNN) algorithm trained to accurately predict the likelihood of a peptide presented by human leukocyte antigen (HLA)-I molecules
We aimed to improve the precision and specificity of the HLApeptide prediction approaches through a novel tool that has been trained on improved training data and a new supervised machine learning model

Summary

Introduction

Cancer develops as a result of the accumulation of tumor-specific somatic mutations [1,2,3], where non-silent mutations in the coding region could be recognized as beacons of “foreign” by the immune system, named neoantigen [4, 5]. They can elicit a protective anti-tumor response when presented on the surface of cancer cells by the major histocompatibility complex (MHC) [ called human leukocyte antigen (HLA)].

Objectives

Methods

Results

Conclusion