INeo-Epp: A Novel T-Cell HLA Class-I Immunogenicity or Neoantigenic Epitope Prediction Method Based on Sequence-Related Amino Acid Features

Anthrax Vaccination, Gulf War Illness, and Human Leukocyte Antigen (HLA)

We report on a highly significant, positive association between anthrax vaccination and occurrence of Gulf War Illness (GWI) in 111 Gulf War veterans (42 with GWI and 69 controls). GWI was diagnosed in 47.1% of vaccinated veterans but only in 17.2% of non-vaccinated veterans (Pearson χ2 = 7.08, p = 0.008; odds ratio = 3.947; relative risk = 2.617), with 1.6x higher GWI symptom severity in vaccinated veterans (p = 0.007, F-test in analysis of covariance). Next, we tested the hypothesis that the susceptibility to GWI following anthrax vaccination could be due to inability to make antibodies against the anthrax protective antigen (PA), the key protein contained in the vaccine. Since the first step in initiating antibody production would be the binding of PA peptide fragments (typically 15-amino acid long [15-mer]) to peptide-binding motifs of human leukocyte antigen (HLA) Class II molecules, we assessed the binding-motif affinities of such HLA specific molecules to all linear 15-mer peptide fragments of the anthrax PA. We identified a total of 58 HLA Class II alleles carried by the veterans in our sample and found that, of those, 18 (31%) were present in the vaccinated group that did not develop GWI but were absent from the vaccinated group who developed GWI. Remarkably, in silico analyses revealed very high binding affinities of peptide-binding motifs of those 18 HLA alleles with fragments of anthrax vaccine PA, leading to the successful production of anti-PA antibodies. Conversely, the absence of these protective HLA alleles points to a reduced ability to develop antibodies against PA, thus resulting in harmful PA persistence and development of GWI.

Artificial intelligence and neoantigens: paving the path for precision cancer immunotherapy

Cancer immunotherapy has witnessed rapid advancement in recent years, with a particular focus on neoantigens as promising targets for personalized treatments. The convergence of immunogenomics, bioinformatics, and artificial intelligence (AI) has propelled the development of innovative neoantigen discovery tools and pipelines. These tools have revolutionized our ability to identify tumor-specific antigens, providing the foundation for precision cancer immunotherapy. AI-driven algorithms can process extensive amounts of data, identify patterns, and make predictions that were once challenging to achieve. However, the integration of AI comes with its own set of challenges, leaving space for further research. With particular focus on the computational approaches, in this article we have explored the current landscape of neoantigen prediction, the fundamental concepts behind, the challenges and their potential solutions providing a comprehensive overview of this rapidly evolving field.

Tumor-Derived Antigenic Peptides as Potential Cancer Vaccines

Peptide antigens derived from tumors have been observed to elicit protective immune responses, categorized as either tumor-associated antigens (TAAs) or tumor-specific antigens (TSAs). Subunit cancer vaccines incorporating these antigens have shown promise in inducing protective immune responses, leading to cancer prevention or eradication. Over recent years, peptide-based cancer vaccines have gained popularity as a treatment modality and are often combined with other forms of cancer therapy. Several clinical trials have explored the safety and efficacy of peptide-based cancer vaccines, with promising outcomes. Advancements in techniques such as whole-exome sequencing, next-generation sequencing, and in silico methods have facilitated the identification of antigens, making it increasingly feasible. Furthermore, the development of novel delivery methods and a deeper understanding of tumor immune evasion mechanisms have heightened the interest in these vaccines among researchers. This article provides an overview of novel insights regarding advancements in the field of peptide-based vaccines as a promising therapeutic avenue for cancer treatment. It summarizes existing computational methods for tumor neoantigen prediction, ongoing clinical trials involving peptide-based cancer vaccines, and recent studies on human vaccination experiments.

Association between brain cancer immunogenetic profile and in silico immunogenicities of 11 viruses

Several viruses including human herpes viruses (HHVs), human polyomavirus JCV, and human papilloma virus (HPV) have been implicated in brain cancer, albeit inconsistently. Since human leukocyte antigen (HLA) is centrally involved in the human immune response to viruses and has been implicated in brain cancer, we evaluated in silico the immunogenicity between 69 Class I HLA alleles with epitopes of proteins of 9 HHVs, JCV, and HPV with respect to a population-based HLA-brain cancer profile. We found that immunogenicity varied widely across HLA alleles with HLA-C alleles exhibiting the highest immunogenicity, and that immunogenicity scores were negatively associated with the population-based HLA-brain cancer profile, particularly for JCV, HHV6A, HHV5, HHV3, HHV8, and HHV7. Consistent with the role of HLA in foreign antigen elimination, the findings suggest that viruses with proteins of high HLA immunogenicity are eliminated more effectively and, consequently, less likely to cause brain cancer; conversely, the absence of highly immunogenic HLA may allow the viral antigens to persist, contributing to cancer.

Neoantigen-targeted TCR-engineered T cell immunotherapy: current advances and challenges

Adoptive cell therapy using T cell receptor-engineered T cells (TCR-T) is a promising approach for cancer therapy with an expectation of no significant side effects. In the human body, mature T cells are armed with an incredible diversity of T cell receptors (TCRs) that theoretically react to the variety of random mutations generated by tumor cells. The outcomes, however, of current clinical trials using TCR-T cell therapies are not very successful especially involving solid tumors. The therapy still faces numerous challenges in the efficient screening of tumor-specific antigens and their cognate TCRs. In this review, we first introduce TCR structure-based antigen recognition and signaling, then describe recent advances in neoantigens and their specific TCR screening technologies, and finally summarize ongoing clinical trials of TCR-T therapies against neoantigens. More importantly, we also present the current challenges of TCR-T cell-based immunotherapies, e.g., the safety of viral vectors, the mismatch of T cell receptor, the impediment of suppressive tumor microenvironment. Finally, we highlight new insights and directions for personalized TCR-T therapy.

Negative association between multiple sclerosis immunogenetic profile and in silico immunogenicities of 12 viruses

Human Leukocyte Antigen (HLA) is involved in both multiple sclerosis (MS) and immune response to viruses. Here we investigated the virus-HLA immunogenicity (V-HLA) of 12 viruses implicated in MS with respect to 17 HLA Class I alleles positively associated to MS prevalence in 14 European countries. Overall, higher V-HLA immunogenicity was associated with smaller MS-HLA effect, with human herpes virus 3 (HHV3), JC human polyoma virus (JCV), HHV1, HHV4, HHV7, HHV5 showing the strongest association, followed by HHV8, HHV6A, and HHV6B (moderate association), and human endogenous retrovirus (HERV-W), HHV2, and human papilloma virus (HPV) (weakest association). These findings suggest that viruses with proteins of high HLA immunogenicity are eliminated more effectively and, consequently, less likely to be involved in MS.

Artificial intelligence applied in neoantigen identification facilitates personalized cancer immunotherapy

The field of cancer neoantigen investigation has developed swiftly in the past decade. Predicting novel and true neoantigens derived from large multi-omics data became difficult but critical challenges. The rise of Artificial Intelligence (AI) or Machine Learning (ML) in biomedicine application has brought benefits to strengthen the current computational pipeline for neoantigen prediction. ML algorithms offer powerful tools to recognize the multidimensional nature of the omics data and therefore extract the key neoantigen features enabling a successful discovery of new neoantigens. The present review aims to outline the significant technology progress of machine learning approaches, especially the newly deep learning tools and pipelines, that were recently applied in neoantigen prediction. In this review article, we summarize the current state-of-the-art tools developed to predict neoantigens. The standard workflow includes calling genetic variants in paired tumor and blood samples, and rating the binding affinity between mutated peptide, MHC (I and II) and T cell receptor (TCR), followed by characterizing the immunogenicity of tumor epitopes. More specifically, we highlight the outstanding feature extraction tools and multi-layer neural network architectures in typical ML models. It is noted that more integrated neoantigen-predicting pipelines are constructed with hybrid or combined ML algorithms instead of conventional machine learning models. In addition, the trends and challenges in further optimizing and integrating the existing pipelines are discussed.

Molecular Bases of Protein Antigenicity and Determinants of Immunogenicity, Anergy, and Mitogenicity

BACKGROUND

The immune system is able to recognize substances that originate from inside or outside the body and are potentially harmful. Foreign substances that bind to immune system components exhibit antigenicity and are defined as antigens. The antigens exhibiting immunogenicity can induce innate or adaptive immune responses and give rise to humoral or cell-mediated immunity. The antigens exhibiting mitogenicity can cross-link cell membrane receptors on B and T lymphocytes leading to cell proliferation. All antigens vary greatly in physicochemical features such as biochemical nature, structural complexity, molecular size, foreignness, solubility, and so on.

OBJECTIVE

Thus, this review aims to describe the molecular bases of protein-antigenicity and those molecular bases that lead to an immune response, lymphocyte proliferation, or unresponsiveness.

CONCLUSION

The epitopes of an antigen are located in surface areas; they are about 880-3,300 Da in size. They are protein, carbohydrate, or lipid in nature. Soluble antigens are smaller than 1 nm and are endocytosed less efficiently than particulate antigens. The more the structural complexity of an antigen increases, the more the antigenicity increases due to the number and variety of epitopes. The smallest immunogens are about 4,000-10,000 Da in size. The more phylogenetically distant immunogens are from the immunogen-recipient, the more immunogenicity increases. Antigens that are immunogens can trigger an innate or adaptive immune response. The innate response is induced by antigens that are pathogen-associated molecular patterns. Exogenous antigens, T Dependent or T Independent, induce humoral immunogenicity. TD protein-antigens require two epitopes, one sequential and one conformational to induce antibodies, whereas, TI non-protein-antigens require only one conformational epitope to induce low-affinity antibodies. Endogenous protein antigens require only one sequential epitope to induce cell-mediated immunogenicity.

CAD v1.0: Cancer Antigens Database Platform for Cancer Antigen Algorithm Development and Information Exploration

Cancer vaccines have gradually attracted attention for their tremendous preclinical and clinical performance. With the development of next-generation sequencing technologies and related algorithms, pipelines based on sequencing and machine learning methods have become mainstream in cancer antigen prediction; of particular focus are neoantigens, mutation peptides that only exist in tumor cells that lack central tolerance and have fewer side effects. The rapid prediction and filtering of neoantigen peptides are crucial to the development of neoantigen-based cancer vaccines. However, due to the lack of verified neoantigen datasets and insufficient research on the properties of neoantigens, neoantigen prediction algorithms still need to be improved. Here, we recruited verified cancer antigen peptides and collected as much relevant peptide information as possible. Then, we discussed the role of each dataset for algorithm improvement in cancer antigen research, especially neoantigen prediction. A platform, Cancer Antigens Database (CAD, http://cad.bio-it.cn/), was designed to facilitate users to perform a complete exploration of cancer antigens online.

dbPepNeo2.0: A Database for Human Tumor Neoantigen Peptides From Mass Spectrometry and TCR Recognition

Neoantigens are widely reported to induce T-cell response and lead to tumor regression, indicating a promising potential to immunotherapy. Previously, we constructed an open-access database, i.e., dbPepNeo, providing a systematic resource for human tumor neoantigens to storage and query. In order to expand data volume and application scope, we updated dbPepNeo to version 2.0 (http://www.biostatistics.online/dbPepNeo2). Here, we provide about 801 high-confidence (HC) neoantigens (increased by 170%) and 842,289 low-confidence (LC) HLA immunopeptidomes (increased by 107%). Notably, 55 class II HC neoantigens and 630 neoantigen-reactive T-cell receptor-β (TCRβ) sequences were firstly included. Besides, two new analytical tools are developed, DeepCNN-Ineo and BLASTdb. DeepCNN-Ineo predicts the immunogenicity of class I neoantigens, and BLASTdb performs local alignments to look for sequence similarities in dbPepNeo2.0. Meanwhile, the web features and interface have been greatly improved and enhanced.

Computational cancer neoantigen prediction: current status and recent advances

Over the last few decades, immunotherapy has shown significant therapeutic efficacy in a broad range of cancer types. Antitumor immune responses are contingent on the recognition of tumor-specific antigens, which are termed neoantigens. Tumor neoantigens are ideal targets for immunotherapy since they can be recognized as non-self antigens by the host immune system and thus are able to elicit an antitumor T-cell response. There are an increasing number of studies that highlight the importance of tumor neoantigens in immunoediting and in the sensitivity to immune checkpoint blockade. Therefore, one of the most fundamental tasks in the field of immuno-oncology research is the identification of patient-specific neoantigens. To this end, a plethora of computational approaches have been developed in order to predict tumor-specific aberrant peptides and quantify their likelihood of binding to patients' human leukocyte antigen molecules in order to be recognized by T cells. In this review, we systematically summarize and present the most recent advances in computational neoantigen prediction, and discuss the challenges and novel methods that are being developed to resolve them.

•

Tumors have the ability to acquire immune escape mechanisms.

•

Tumor-specific aberrant peptides (neoantigens) can elicit an immune response by the host immune system.

•

The identification of neoantigens is one of the most fundamental tasks in the field of immuno-oncology research.

•

A plethora of computational approaches have been developed in order to predict patient-specificneoantigens.

DeepImmuno: deep learning-empowered prediction and generation of immunogenic peptides for T-cell immunity

Cytolytic T-cells play an essential role in the adaptive immune system by seeking out, binding and killing cells that present foreign antigens on their surface. An improved understanding of T-cell immunity will greatly aid in the development of new cancer immunotherapies and vaccines for life-threatening pathogens. Central to the design of such targeted therapies are computational methods to predict non-native peptides to elicit a T-cell response, however, we currently lack accurate immunogenicity inference methods. Another challenge is the ability to accurately simulate immunogenic peptides for specific human leukocyte antigen alleles, for both synthetic biological applications, and to augment real training datasets. Here, we propose a beta-binomial distribution approach to derive peptide immunogenic potential from sequence alone. We conducted systematic benchmarking of five traditional machine learning (ElasticNet, K-nearest neighbors, support vector machine, Random Forest and AdaBoost) and three deep learning models (convolutional neural network (CNN), Residual Net and graph neural network) using three independent prior validated immunogenic peptide collections (dengue virus, cancer neoantigen and SARS-CoV-2). We chose the CNN as the best prediction model, based on its adaptivity for small and large datasets and performance relative to existing methods. In addition to outperforming two highly used immunogenicity prediction algorithms, DeepImmuno-CNN correctly predicts which residues are most important for T-cell antigen recognition and predicts novel impacts of SARS-CoV-2 variants. Our independent generative adversarial network (GAN) approach, DeepImmuno-GAN, was further able to accurately simulate immunogenic peptides with physicochemical properties and immunogenicity predictions similar to that of real antigens. We provide DeepImmuno-CNN as source code and an easy-to-use web interface.

Recent progress on MHC-I epitope prediction in tumor immunotherapy

Tumor immunotherapy has now become one of the most potential therapy for those intractable cancer diseases. The antigens on the cancer cell surfaces are the keys for the immune system to recognize and eliminate them. As reported, the immunogenicity of the tumor antigens could be determined by the binding between the key epitope peptides and MHC molecules. In recent years, the approaches to anticipate the peptides from the candidate epitopes have gradually changed into more efficient methods. Including the improved conventional methods, more diverse methods were coming into view. Here we review the anticipated methods of the tumor associated epitopes that specifically bind with major histocompatibility complex (MHC) class I molecules, and the recent advances and applications of those epitope prediction methods.

DeepImmuno: Deep learning-empowered prediction and generation of immunogenic peptides for T cell immunity

T-cells play an essential role in the adaptive immune system by seeking out, binding and destroying foreign antigens presented on the cell surface of diseased cells. An improved understanding of T-cell immunity will greatly aid in the development of new cancer immunotherapies and vaccines for life threatening pathogens. Central to the design of such targeted therapies are computational methods to predict non-native epitopes to elicit a T cell response, however, we currently lack accurate immunogenicity inference methods. Another challenge is the ability to accurately simulate immunogenic peptides for specific human leukocyte antigen (HLA) alleles, for both synthetic biological applications and to augment real training datasets. Here, we proposed a beta-binomial distribution approach to derive epitope immunogenic potential from sequence alone. We conducted systematic benchmarking of five traditional machine learning (ElasticNet, KNN, SVM, Random Forest, AdaBoost) and three deep learning models (CNN, ResNet, GNN) using three independent prior validated immunogenic peptide collections (dengue virus, cancer neoantigen and SARS-Cov-2). We chose the CNN model as the best prediction model based on its adaptivity for small and large datasets, and performance relative to existing methods. In addition to outperforming two highly used immunogenicity prediction algorithms, DeepHLApan and IEDB, DeepImmuno-CNN further correctly predicts which residues are most important for T cell antigen recognition. Our independent generative adversarial network (GAN) approach, DeepImmuno-GAN, was further able to accurately simulate immunogenic peptides with physiochemical properties and immunogenicity predictions similar to that of real antigens. We provide DeepImmuno-CNN as source code and an easy-to-use web interface.

DATA AVAILABILITY

DeepImmuno Python3 code is available at https://github.com/frankligy/DeepImmuno . The DeepImmuno web portal is available from https://deepimmuno.herokuapp.com . The data in this article is available in GitHub and supplementary materials.

Main Strategies for the Identification of Neoantigens

Genetic instability of tumors leads to the appearance of numerous tumor-specific somatic mutations that could potentially result in the production of mutated peptides that are presented on the cell surface by the MHC molecules. Peptides of this kind are commonly called neoantigens. Their presence on the cell surface specifically distinguishes tumors from healthy tissues. This feature makes neoantigens a promising target for immunotherapy. The rapid evolution of high-throughput genomics and proteomics makes it possible to implement these techniques in clinical practice. In particular, they provide useful tools for the investigation of neoantigens. The most valuable genomic approach to this problem is whole-exome sequencing coupled with RNA-seq. High-throughput mass-spectrometry is another option for direct identification of MHC-bound peptides, which is capable of revealing the entire MHC-bound peptidome. Finally, structure-based predictions could significantly improve the understanding of physicochemical and structural features that affect the immunogenicity of peptides. The development of pipelines combining such tools could improve the accuracy of the peptide selection process and decrease the required time. Here we present a review of the main existing approaches to investigating the neoantigens and suggest a possible ideal pipeline that takes into account all modern trends in the context of neoantigen discovery.