Predicting MHC class I binder: existing approaches and a novel recurrent neural network solution

Comprehensive Profiling of Computational Techniques for Sequencing-Based HLA Immune Signatures Extraction

HLA typing is crucial for clinical and research applications, including transplantation, disease association studies and personalised medicine. This study provides an in-depth analysis of five key challenges in computational HLA typing methods: varying lengths, high polymorphism, complex phylogenetic structures, high sequence similarity and the requirement for frequent updates. This study evaluates 27 computation-based HLA typing tools developed over the past 12 years, using a novel sequencing dataset of A549 cell lines with matched short-read RNA-Seq and long-read Iso-Seq data. We comprehensively investigated these 27 tools in terms of accessibility, capability, reliability, reproducibility, scalability and performance. In addition, we discuss the advantages and disadvantages of current tools and identify critical areas for future research and development to advance HLA typing technologies.

Machine learning application to predict binding affinity between peptide containing non-canonical amino acids and HLA0201

Class 1 major histocompatibility complexes (MHC-I), encoded by the highly polymorphic HLA-A, HLA-B, and HLA-C genes in humans, are expressed on all nucleated cells. Both self and foreign proteins are processed to peptides of 8 to 10 amino acids, loaded into MCH-1 within the endoplasmic reticulum and then presented on the cell surface. Foreign peptides presented in this fashion activate CD8+ T cells and their immunogenicity correlates with their affinity for the MHC-1 binding groove. Thus, predicting antigen binding affinity for MHC-I is a valuable tool for identifying potentially immunogenic antigens. While quite a few predictors for MHC-I binding exist, there are no currently available tools that can predict antigen/MHC-I binding affinity for antigens with explicitly labeled post-translational modifications or unusual/non-canonical amino acids (NCAAs). However, such modifications are increasingly recognized as critical mediators of peptide immunogenicity. In this work, we propose a machine learning application that quantifies the binding affinity of epitopes containing NCAAs to MHC-I and compares its performance with other commonly used regressors. Our model demonstrates robust performance, with 5-fold cross-validation yielding an R2 value of 0.477 and a root-mean-square error (RMSE) of 0.735, indicating strong predictive capability for peptides with NCAAs. This work provides a valuable tool for the computational design and optimization of peptides incorporating NCAAs, potentially accelerating the development of novel peptide-based therapeutics with enhanced properties and efficacy.

StackIL10: A stacking ensemble model for the improved prediction of IL-10 inducing peptides

Interleukin-10, a highly effective cytokine recognized for its anti-inflammatory properties, plays a critical role in the immune system. In addition to its well-documented capacity to mitigate inflammation, IL-10 can unexpectedly demonstrate pro-inflammatory characteristics under specific circumstances. The presence of both aspects emphasizes the vital need to identify the IL-10-induced peptide. To mitigate the drawbacks of manual identification, which include its high cost, this study introduces StackIL10, an ensemble learning model based on stacking, to identify IL-10-inducing peptides in a precise and efficient manner. Ten Amino-acid-composition-based Feature Extraction approaches are considered. The StackIL10, stacking ensemble, the model with five optimized Machine Learning Algorithm (specifically LGBM, RF, SVM, Decision Tree, KNN) as the base learners and a Logistic Regression as the meta learner was constructed, and the identification rate reached 91.7%, MCC of 0.833 with 0.9078 Specificity. Experiments were conducted to examine the impact of various enhancement techniques on the correctness of IL-10 Prediction. These experiments included comparisons between single models and various combinations of stacking-based ensemble models. It was demonstrated that the model proposed in this study was more effective than singular models and produced satisfactory results, thereby improving the identification of peptides that induce IL-10.

Enhancing antigenic peptide discovery: Improved MHC-I binding prediction and methodology

The Major Histocompatibility Complex (MHC) is a critical element of the vertebrate cellular immune system, responsible for presenting peptides derived from intracellular proteins. MHC-I presentation is pivotal in the immune response and holds considerable potential in the realms of vaccine development and cancer immunotherapy. This study delves into the limitations of current methods and benchmarks for MHC-I presentation. We introduce a novel benchmark designed to assess generalization properties and the reliability of models on unseen MHC molecules and peptides, with a focus on the Human Leukocyte Antigen (HLA)-a specific subset of MHC genes present in humans. Finally, we introduce HLABERT, a pretrained language model that outperforms previous methods significantly on our benchmark and establishes a new state-of-the-art on existing benchmarks.

ConvNeXt-MHC: improving MHC-peptide affinity prediction by structure-derived degenerate coding and the ConvNeXt model

Peptide binding to major histocompatibility complex (MHC) proteins plays a critical role in T-cell recognition and the specificity of the immune response. Experimental validation such peptides is extremely resource-intensive. As a result, accurate computational prediction of binding peptides is highly important, particularly in the context of cancer immunotherapy applications, such as the identification of neoantigens. In recent years, there is a significant need to continually improve the existing prediction methods to meet the demands of this field. We developed ConvNeXt-MHC, a method for predicting MHC-I-peptide binding affinity. It introduces a degenerate encoding approach to enhance well-established panspecific methods and integrates transfer learning and semi-supervised learning methods into the cutting-edge deep learning framework ConvNeXt. Comprehensive benchmark results demonstrate that ConvNeXt-MHC outperforms state-of-the-art methods in terms of accuracy. We expect that ConvNeXt-MHC will help us foster new discoveries in the field of immunoinformatics in the distant future. We constructed a user-friendly website at http://www.combio-lezhang.online/predict/, where users can access our data and application.

A comprehensive analysis of the IEDB MHC class-I automated benchmark

In 2014, the Immune Epitope Database automated benchmark was created to compare the performance of the MHC class I binding predictors. However, this is not a straightforward process due to the different and non-standardized outputs of the methods. Additionally, some methods are more restrictive regarding the HLA alleles and epitope sizes for which they predict binding affinities, while others are more comprehensive. To address how these problems impacted the ranking of the predictors, we developed an approach to assess the reliability of different metrics. We found that using percentile-ranked results improved the stability of the ranks and allowed the predictors to be reliably ranked despite not being evaluated on the same data. We also found that given the rate new data are incorporated into the benchmark, a new method must wait for at least 4 years to be ranked against the pre-existing methods. The best-performing tools with statistically indistinguishable scores in this benchmark were NetMHCcons, NetMHCpan4.0, ANN3.4, NetMHCpan3.0 and NetMHCpan2.8. The results of this study will be used to improve the evaluation and display of benchmark performance. We highly encourage anyone working on MHC binding predictions to participate in this benchmark to get an unbiased evaluation of their predictors.

Prediction of Major Histocompatibility Complex Binding with Bilateral and Variable Long Short Term Memory Networks

Simple Summary

Major histocompatibility complex molecules are of significant biological and clinical importance due to their utility in immunotherapy. The prediction of potential MHC binding peptides can estimate a T-cell immune response. The variable length of existing MHC binding peptides creates difficulty for MHC binding prediction algorithms. Thus, we utilized a bilateral and variable long-short term memory neural network to address this specific problem and developed a novel MHC binding prediction tool.

Abstract

As an important part of immune surveillance, major histocompatibility complex (MHC) is a set of proteins that recognize foreign molecules. Computational prediction methods for MHC binding peptides have been developed. However, existing methods share the limitation of fixed peptide sequence length, which necessitates the training of models by peptide length or prediction with a length reduction technique. Using a bidirectional long short-term memory neural network, we constructed BVMHC, an MHC class I and II binding prediction tool that is independent of peptide length. The performance of BVMHC was compared to seven MHC class I prediction tools and three MHC class II prediction tools using eight performance criteria independently. BVMHC attained the best performance in three of the eight criteria for MHC class I, and the best performance in four of the eight criteria for MHC class II, including accuracy and AUC. Furthermore, models for non-human species were also trained using the same strategy and made available for applications in mice, chimpanzees, macaques, and rats. BVMHC is composed of a series of peptide length independent MHC class I and II binding predictors. Models from this study have been implemented in an online web portal for easy access and use.

PredMHC: An Effective Predictor of Major Histocompatibility Complex Using Mixed Features

The major histocompatibility complex (MHC) is a large locus on vertebrate DNA that contains a tightly linked set of polymorphic genes encoding cell surface proteins essential for the adaptive immune system. The groups of proteins encoded in the MHC play an important role in the adaptive immune system. Therefore, the accurate identification of the MHC is necessary to understand its role in the adaptive immune system. An effective predictor called PredMHC is established in this study to identify the MHC from protein sequences. Firstly, PredMHC encoded a protein sequence with mixed features including 188D, APAAC, KSCTriad, CKSAAGP, and PAAC. Secondly, three classifiers including SGD, SMO, and random forest were trained on the mixed features of the protein sequence. Finally, the prediction result was obtained by the voting of the three classifiers. The experimental results of the 10-fold cross-validation test in the training dataset showed that PredMHC can obtain 91.69% accuracy. Experimental results on comparison with other features, classifiers, and existing methods showed the effectiveness of PredMHC in predicting the MHC.