Deciphering HLA-I motifs across HLA peptidomes improves neo-antigen predictions and identifies allostery regulating HLA specificity

Toward equitable major histocompatibility complex binding predictions

Deep learning tools that predict peptide binding by major histocompatibility complex (MHC) proteins play an essential role in developing personalized cancer immunotherapies and vaccines. In order to ensure equitable health outcomes from their application, MHC binding prediction methods must work well across the vast landscape of MHC alleles observed across human populations. Here, we show that there are alarming disparities across individuals in different racial and ethnic groups in how much binding data are associated with their MHC alleles. We introduce a machine learning framework to assess the impact of this data imbalance for predicting binding for any given MHC allele, and apply it to develop a state-of-the-art MHC binding prediction model that additionally provides per-allele performance estimates. We demonstrate that our MHC binding model successfully mitigates much of the data disparities observed across racial groups. To address remaining inequities, we devise an algorithmic strategy for targeted data collection. Our work lays the foundation for further development of equitable MHC binding models for use in personalized immunotherapies.

Leveraging mRNA technology for antigen based immuno-oncology therapies

The application of messenger RNA (mRNA) technology in antigen-based immuno-oncology therapies represents a significant advancement in cancer treatment. Cancer vaccines are an effective combinatorial partner to sensitize the host immune system to the tumor and boost the efficacy of immune therapies. Selecting suitable tumor antigens is the key step to devising effective vaccinations and amplifying the immune response. Tumor neoantigens are de novo epitopes derived from somatic mutations, avoiding T-cell central tolerance of self-epitopes and inducing immune responses to tumors. The identification and prioritization of patient-specific tumor neoantigens are based on advanced computational algorithms taking advantage of the profiling with next-generation sequencing considering factors involved in human leukocyte antigen (HLA)-peptide-T-cell receptor (TCR) complex formation, including peptide presentation, HLA-peptide affinity, and TCR recognition. This review discusses the development and clinical application of mRNA vaccines in oncology, with a particular focus on recent clinical trials and the computational workflows and methodologies for identifying both shared and individual antigens. While this review centers on therapeutic mRNA vaccines targeting existing tumors, it does not cover preventative vaccines. Preclinical experimental validations are crucial in cancer vaccine development, but we emphasize the computational approaches that facilitate neoantigen selection and design, highlighting their role in advancing mRNA vaccine development. The versatility and rapid development potential of mRNA make it an ideal platform for personalized neoantigen immunotherapy. We explore various strategies for antigen target identification, including tumor-associated and tumor-specific antigens and the computational tools used to predict epitopes capable of eliciting strong immune responses. We address key design considerations for enhancing the immunogenicity and stability of mRNA vaccines, as well as emerging trends and challenges in the field. This comprehensive overview highlights the therapeutic potential of mRNA-based cancer vaccines and underscores ongoing research efforts aimed at optimizing these therapies for improved clinical outcomes.

HLA I immunopeptidome of synthetic long peptide pulsed human dendritic cells for therapeutic vaccine design

Synthetic long peptides (SLPs) are a promising vaccine modality that exploit dendritic cells (DC) to treat chronic infections or cancer. Currently, the design of SLPs relies on in silico prediction and multifactorial T cells assays to determine which SLPs are best cross-presented on DC human leukocyte antigen class I (HLA-I). Furthermore, it is unknown how TLR ligand-based adjuvants affect DC cross-presentation. Here, we generated a unique, high-quality immunopeptidome dataset of human DCs pulsed with 12 hepatitis B virus (HBV)-based SLPs combined with either a TLR1/2 (Amplivant®) or TLR3 (PolyI:C) ligand. The obtained immunopeptidome reflected adjuvant-induced differences, but no differences in cross-presentation of SLPs. We uncovered dominant (cross-)presentation on B-alleles, and identified 33 unique SLP-derived HLA-I peptides, several of which were not in silico predicted and some were consistently found across donors. Our work puts forward DC immunopeptidomics as a valuable tool for therapeutic vaccine design.

Emerging biochemical, microbial and immunological evidence in the search for why HLA-B∗27 confers risk for spondyloarthritis

The strong association of the human leukocyte antigen B∗27 alleles (HLA-B∗27) with spondyloarthritis and related rheumatic conditions has long fascinated researchers, yet the precise mechanisms underlying its pathogenicity remain elusive. Here, we review how interplay between the microbiome, the immune system, and the enigmatic HLA-B∗27 could trigger spondyloarthritis, with a focus on whether HLA-B∗27 presents an arthritogenic peptide. We propose mechanisms by which the unique biochemical characteristics of the HLA-B∗27 protein structure, particularly its peptide binding groove, could dictate its propensity to induce pathological T cell responses. We further provide new insights into how TRBV9+ CD8+ T cells are implicated in the disease process, as well as how the immunometabolism of T cells modulates tissue-specific inflammatory responses in spondyloarthritis. Finally, we present testable models and suggest approaches to this problem in future studies given recent advances in computational biology, chemical biology, structural biology, and small-molecule therapeutics.

Establishment of a novel tumor neoantigen prediction tool for personalized vaccine design

The personalized neoantigen nanovaccine (PNVAC) platform for patients with gastric cancer we established previously exhibited promising anti-tumor immunoreaction. However, limited by the ability of traditional neoantigen prediction tools, a portion of epitopes failed to induce specific immune response. In order to filter out more neoantigens to optimize our PNVAC platform, we develop a novel neoantigen prediction model, NUCC. This prediction tool trained through a deep learning approach exhibits better neoantigen prediction performance than other prediction tools, not only in two independent epitope datasets, but also in a totally new epitope dataset we construct from scratch, including 25 patients with advance gastric cancer and 150 candidate mutant peptides, 13 of which prove to be neoantigen by immunogenicity test in vitro. Our work lay the foundation for the improvement of our PNVAC platform for gastric cancer in the future.

Recent progress in mRNA cancer vaccines

The research and development of messenger RNA (mRNA) cancer vaccines have gradually overcome numerous challenges through the application of personalized cancer antigens, structural optimization of mRNA, and the development of alternative RNA-based vectors and efficient targeted delivery vectors. Clinical trials are currently underway for various cancer vaccines that encode tumor-associated antigens (TAAs), tumor-specific antigens (TSAs), or immunomodulators. In this paper, we summarize the optimization of mRNA and the emergence of RNA-based expression vectors in cancer vaccines. We begin by reviewing the advancement and utilization of state-of-the-art targeted lipid nanoparticles (LNPs), followed by presenting the primary classifications and clinical applications of mRNA cancer vaccines. Collectively, mRNA vaccines are emerging as a central focus in cancer immunotherapy, offering the potential to address multiple challenges in cancer treatment, either as standalone therapies or in combination with current cancer treatments.

Towards designing improved cancer immunotherapy targets with a peptide-MHC-I presentation model, HLApollo

Based on the success of cancer immunotherapy, personalized cancer vaccines have emerged as a leading oncology treatment. Antigen presentation on MHC class I (MHC-I) is crucial for the adaptive immune response to cancer cells, necessitating highly predictive computational methods to model this phenomenon. Here, we introduce HLApollo, a transformer-based model for peptide-MHC-I (pMHC-I) presentation prediction, leveraging the language of peptides, MHC, and source proteins. HLApollo provides end-to-end treatment of MHC-I sequences and deconvolution of multi-allelic data, using a negative-set switching strategy to mitigate misassigned negatives in unlabelled ligandome data. HLApollo shows a 12.65% increase in average precision (AP) on ligandome data and a 4.1% AP increase on immunogenicity test data compared to next-best models. Incorporating protein features from protein language models yields further gains and reduces the need for gene expression measurements. Guided by clinical use, we demonstrate pan-allelic generalization which effectively captures rare alleles in underrepresented ancestries.

Cracking the code of adaptive immunity: The role of computational tools

In recent years, the advances in high-throughput and deep sequencing have generated a diverse amount of adaptive immune repertoire data. This surge in data has seen a proportional increase in computational methods aimed to characterize T cell receptor (TCR) repertoires. In this perspective, we will provide a brief commentary on the various domains of TCR repertoire analysis, their respective computational methods, and the ongoing challenges. Given the breadth of methods and applications of TCR analysis, we will focus our perspective on sequence-based computational methods.

Advances of computational methods enhance the development of multi-epitope vaccines

Vaccine development is one of the most promising fields, and multi-epitope vaccine, which does not need laborious culture processes, is an attractive alternative to classical vaccines with the advantage of safety, and efficiency. The rapid development of algorithms and the accumulation of immune data have facilitated the advancement of computer-aided vaccine design. Here we systemically reviewed the in silico data and algorithms resource, for different steps of computational vaccine design, including immunogen selection, epitope prediction, vaccine construction, optimization, and evaluation. The performance of different available tools on epitope prediction and immunogenicity evaluation was tested and compared on benchmark datasets. Finally, we discuss the future research direction for the construction of a multiepitope vaccine.

Meta learning for mutant HLA class I epitope immunogenicity prediction to accelerate cancer clinical immunotherapy

Accurate prediction of binding between human leukocyte antigen (HLA) class I molecules and antigenic peptide segments is a challenging task and a key bottleneck in personalized immunotherapy for cancer. Although existing prediction tools have demonstrated significant results using established datasets, most can only predict the binding affinity of antigenic peptides to HLA and do not enable the immunogenic interpretation of new antigenic epitopes. This limitation results from the training data for the computational models relying heavily on a large amount of peptide-HLA (pHLA) eluting ligand data, in which most of the candidate epitopes lack immunogenicity. Here, we propose an adaptive immunogenicity prediction model, named MHLAPre, which is trained on the large-scale MS-derived HLA I eluted ligandome (mostly presented by epitopes) that are immunogenic. Allele-specific and pan-allelic prediction models are also provided for endogenous peptide presentation. Using a meta-learning strategy, MHLAPre rapidly assessed HLA class I peptide affinities across the whole pHLA pairs and accurately identified tumor-associated endogenous antigens. During the process of adaptive immune response of T-cells, pHLA-specific binding in the antigen presentation is only a pre-task for CD8+ T-cell recognition. The key factor in activating the immune response is the interaction between pHLA complexes and T-cell receptors (TCRs). Therefore, we performed transfer learning on the pHLA model using the pHLA-TCR dataset. In pHLA binding task, MHLAPre demonstrated significant improvement in identifying neoepitope immunogenicity compared with five state-of-the-art models, proving its effectiveness and robustness. After transfer learning of the pHLA-TCR data, MHLAPre also exhibited relatively superior performance in revealing the mechanism of immunotherapy. MHLAPre is a powerful tool to identify neoepitopes that can interact with TCR and induce immune responses. We believe that the proposed method will greatly contribute to clinical immunotherapy, such as anti-tumor immunity, tumor-specific T-cell engineering, and personalized tumor vaccine.

HLA-DR4Pred2: An improved method for predicting HLA-DRB1*04:01 binders

HLA-DRB1*04:01 is associated with numerous diseases, including sclerosis, arthritis, diabetes, and COVID-19, emphasizing the need to scan for binders in the antigens to develop immunotherapies and vaccines. Current prediction methods are often limited by their reliance on the small datasets. This study presents HLA-DR4Pred2, developed on a large dataset containing 12,676 binders and an equal number of non-binders. It's an improved version of HLA-DR4Pred, which was trained on a small dataset, containing 576 binders and an equal number of non-binders. All models were trained, optimized, and tested on 80 % of the data using five-fold cross-validation and evaluated on the remaining 20 %. A range of machine learning techniques was employed, achieving maximum AUROC of 0.90 and 0.87, using composition and binary profile features, respectively. The performance of the composition-based model increased to 0.93, when combined with BLAST search. Additionally, models developed on the realistic dataset containing 12,676 binders and 86,300 non-binders, achieved a maximum AUROC of 0.99. Our proposed method outperformed existing methods when we compared the performance of our best model to that of existing methods on the independent dataset. Finally, we developed a standalone tool and a webserver for HLADR4Pred2, enabling the prediction, design, and virtual scanning of HLA-DRB1*04:01 binding peptides, and we also released a Python package available on the Python Package Index (https://webs.iiitd.edu.in/raghava/hladr4pred2/; https://github.com/raghavagps/hladr4pred2; https://pypi.org/project/hladr4pred2/).

ImmuneApp for HLA-I epitope prediction and immunopeptidome analysis

Advances in mass spectrometry accelerates the characterization of HLA ligandome, necessitating the development of efficient methods for immunopeptidomics analysis and (neo)antigen prediction. We develop ImmuneApp, an interpretable deep learning framework trained on extensive HLA ligand datasets, which improves the prediction of HLA-I epitopes, prioritizes neoepitopes, and enhances immunopeptidomics deconvolution. ImmuneApp extracts informative embeddings and identifies key residues for pHLA binding. We also present a more accurate model-based deconvolution approach and systematically analyzed 216 multi-allelic immunopeptidomics samples, identifying 835,551 ligands restricted to over 100 HLA-I alleles. Our investigation reveals the effectiveness of the composite model, denoted as ImmuneApp-MA, which integrates mono- and multi-allelic data to enhance predictive performance. Leveraging ImmuneApp-MA as a pre-trained model, we built ImmuneApp-Neo, an immunogenicity predictor that outperforms existing methods for prioritizing immunogenic neoepitope. ImmuneApp demonstrates its utility across various immunopeptidomics datasets, which will promote the discovery of novel neoantigens and the development of new immunotherapies.

Advances in Vaccines for Melanoma

Personalized neoantigen vaccines have achieved major advancements in recent years, with studies in melanoma leading progress in the field. Early clinical trials have demonstrated their feasibility, safety, immunogenicity, and potential efficacy. Advances in sequencing technologies and neoantigen prediction algorithms have substantively improved the identification and prioritization of neoantigens. Innovative delivery platforms now support the rapid and flexible production of vaccines. Several ongoing efforts in the field are aimed at improving the integration of large datasets, refining the training of prediction models, and ensuring the functional validation of vaccine immunogenicity.

3D genome contributes to MHC-II neoantigen prediction

Reliable and ultra-fast DNA and RNA sequencing have been achieved with the emergence of high-throughput sequencing technology. When combining the results of DNA and RNA sequencing for tumor cells of cancer patients, neoantigens that potentially stimulate the immune response of either CD4+ or CD8+ T cells can be identified. However, due to the abundance of somatic mutations and the high polymorphic nature of human leukocyte antigen (HLA) it is challenging to accurately predict the neoantigens. Moreover, comparing to HLA-I presented peptides, the HLA-II presented peptides are more variable in length, making the prediction of HLA-II loaded neoantigens even harder. A number of computational approaches have been proposed to address this issue but none of them considers the DNA origin of the neoantigens from the perspective of 3D genome. Here we investigate the DNA origins of the immune-positive and non-negative HLA-II neoantigens in the context of 3D genome and discovered that the chromatin 3D architecture plays an important role in more effective HLA-II neoantigen prediction. We believe that the 3D genome information will help to increase the precision of HLA-II neoantigen discovery and eventually benefit precision and personalized medicine in cancer immunotherapy.

NeoaPred: a deep-learning framework for predicting immunogenic neoantigen based on surface and structural features of peptide-human leukocyte antigen complexes

MOTIVATION

Neoantigens, derived from somatic mutations in cancer cells, can elicit anti-tumor immune responses when presented to autologous T cells by human leukocyte antigen. Identifying immunogenic neoantigens is crucial for cancer immunotherapy development. However, the accuracy of current bioinformatic methods remains unsatisfactory. Surface and structural features of peptide-HLA class I (pHLA-I) complexes offer valuable insight into the immunogenicity of neoantigens.

RESULTS

We present NeoaPred, a deep-learning framework for neoantigen prediction. NeoaPred accurately constructs pHLA-I complex structures, with 82.37% of the predicted structures showing an RMSD of < 1 Å. Using these structures, NeoaPred integrates differences in surface, structural, and atom group features between the mutant peptide and its wild-type counterpart to predict a foreignness score. This foreignness score is an effective factor for neoantigen prediction, achieving an AUROC (Area Under the Receiver Operating Characteristic Curve) of 0.81 and an AUPRC (Area Under the Precision-Recall Curve) of 0.54 in the test set, outperforming existing methods.

AVAILABILITY AND IMPLEMENTATION

The source code is released under an Apache v2.0 license and is available at the GitHub repository (https://github.com/Dulab2020/NeoaPred).

optiPRM: A Targeted Immunopeptidomics LC-MS Workflow With Ultra-High Sensitivity for the Detection of Mutation-Derived Tumor Neoepitopes From Limited Input Material

Personalized cancer immunotherapies such as therapeutic vaccines and adoptive transfer of T cell receptor-transgenic T cells rely on the presentation of tumor-specific peptides by human leukocyte antigen class I molecules to cytotoxic T cells. Such neoepitopes can for example arise from somatic mutations and their identification is crucial for the rational design of new therapeutic interventions. Liquid chromatography mass spectrometry (LC-MS)-based immunopeptidomics is the only method to directly prove actual peptide presentation and we have developed a parameter optimization workflow to tune targeted assays for maximum detection sensitivity on a per peptide basis, termed optiPRM. Optimization of collision energy using optiPRM allows for the improved detection of low abundant peptides that are very hard to detect using standard parameters. Applying this to immunopeptidomics, we detected a neoepitope in a patient-derived xenograft from as little as 2.5 × 106 cells input. Application of the workflow on small patient tumor samples allowed for the detection of five mutation-derived neoepitopes in three patients. One neoepitope was confirmed to be recognized by patient T cells. In conclusion, optiPRM, a targeted MS workflow reaching ultra-high sensitivity by per peptide parameter optimization, makes the identification of actionable neoepitopes possible from sample sizes usually available in the clinic.

APEX-pHLA: A novel method for accurate prediction of the binding between exogenous short peptides and HLA class I molecules

Human leukocyte antigen (HLA) molecules play critically significant role within the realm of immunotherapy due to their capacities to recognize and bind exogenous antigens such as peptides, subsequently delivering them to immune cells. Predicting the binding between peptides and HLA molecules (pHLA) can expedite the screening of immunogenic peptides and facilitate vaccine design. However, traditional experimental methods are time-consuming and inefficient. In this study, an efficient method based on deep learning was developed for predicting peptide-HLA binding, which treated peptide sequences as linguistic entities. It combined the architectures of textCNN and BiLSTM to create a deep neural network model called APEX-pHLA. This model operated without limitations related to HLA class I allele variants and peptide segment lengths, enabling efficient encoding of sequence features for both HLA and peptide segments. On the independent test set, the model achieved Accuracy, ROC_AUC, F1, and MCC is 0.9449, 0.9850, 0.9453, and 0.8899, respectively. Similarly, on an external test set, the results were 0.9803, 0.9574, 0.8835, and 0.7863, respectively. These findings outperformed fifteen methods previously reported in the literature. The accurate prediction capability of the APEX-pHLA model in peptide-HLA binding might provide valuable insights for future HLA vaccine design.

GIHP: Graph convolutional neural network based interpretable pan-specific HLA-peptide binding affinity prediction

Accurately predicting the binding affinities between Human Leukocyte Antigen (HLA) molecules and peptides is a crucial step in understanding the adaptive immune response. This knowledge can have important implications for the development of effective vaccines and the design of targeted immunotherapies. Existing sequence-based methods are insufficient to capture the structure information. Besides, the current methods lack model interpretability, which hinder revealing the key binding amino acids between the two molecules. To address these limitations, we proposed an interpretable graph convolutional neural network (GCNN) based prediction method named GIHP. Considering the size differences between HLA and short peptides, GIHP represent HLA structure as amino acid-level graph while represent peptide SMILE string as atom-level graph. For interpretation, we design a novel visual explanation method, gradient weighted activation mapping (Grad-WAM), for identifying key binding residues. GIHP achieved better prediction accuracy than state-of-the-art methods across various datasets. According to current research findings, key HLA-peptide binding residues mutations directly impact immunotherapy efficacy. Therefore, we verified those highlighted key residues to see whether they can significantly distinguish immunotherapy patient groups. We have verified that the identified functional residues can successfully separate patient survival groups across breast, bladder, and pan-cancer datasets. Results demonstrate that GIHP improves the accuracy and interpretation capabilities of HLA-peptide prediction, and the findings of this study can be used to guide personalized cancer immunotherapy treatment. Codes and datasets are publicly accessible at: https://github.com/sdustSu/GIHP.

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.

Sliding Window INteraction Grammar (SWING): a generalized interaction language model for peptide and protein interactions

The explosion of sequence data has allowed the rapid growth of protein language models (pLMs). pLMs have now been employed in many frameworks including variant-effect and peptide-specificity prediction. Traditionally, for protein-protein or peptide-protein interactions (PPIs), corresponding sequences are either co-embedded followed by post-hoc integration or the sequences are concatenated prior to embedding. Interestingly, no method utilizes a language representation of the interaction itself. We developed an interaction LM (iLM), which uses a novel language to represent interactions between protein/peptide sequences. Sliding Window Interaction Grammar (SWING) leverages differences in amino acid properties to generate an interaction vocabulary. This vocabulary is the input into a LM followed by a supervised prediction step where the LM's representations are used as features. SWING was first applied to predicting peptide:MHC (pMHC) interactions. SWING was not only successful at generating Class I and Class II models that have comparable prediction to state-of-the-art approaches, but the unique Mixed Class model was also successful at jointly predicting both classes. Further, the SWING model trained only on Class I alleles was predictive for Class II, a complex prediction task not attempted by any existing approach. For de novo data, using only Class I or Class II data, SWING also accurately predicted Class II pMHC interactions in murine models of SLE (MRL/lpr model) and T1D (NOD model), that were validated experimentally. To further evaluate SWING's generalizability, we tested its ability to predict the disruption of specific protein-protein interactions by missense mutations. Although modern methods like AlphaMissense and ESM1b can predict interfaces and variant effects/pathogenicity per mutation, they are unable to predict interaction-specific disruptions. SWING was successful at accurately predicting the impact of both Mendelian mutations and population variants on PPIs. This is the first generalizable approach that can accurately predict interaction-specific disruptions by missense mutations with only sequence information. Overall, SWING is a first-in-class generalizable zero-shot iLM that learns the language of PPIs.

The electrostatic landscape of MHC-peptide binding revealed using inception networks

Predictive modeling of macromolecular recognition and protein-protein complementarity represents one of the cornerstones of biophysical sciences. However, such models are often hindered by the combinatorial complexity of interactions at the molecular interfaces. Exemplary of this problem is peptide presentation by the highly polymorphic major histocompatibility complex class I (MHC-I) molecule, a principal component of immune recognition. We developed human leukocyte antigen (HLA)-Inception, a deep biophysical convolutional neural network, which integrates molecular electrostatics to capture non-bonded interactions for predicting peptide binding motifs across 5,821 MHC-I alleles. These predictions of generated motifs correlate strongly with experimental peptide binding and presentation data. Beyond molecular interactions, the study demonstrates the application of predicted motifs in analyzing MHC-I allele associations with HIV disease progression and patient response to immune checkpoint inhibitors. A record of this paper's transparent peer review process is included in the supplemental information.

Machine learning-based peptide-spectrum match rescoring opens up the immunopeptidome

Immunopeptidomics is a key technology in the discovery of targets for immunotherapy and vaccine development. However, identifying immunopeptides remains challenging due to their non-tryptic nature, which results in distinct spectral characteristics. Moreover, the absence of strict digestion rules leads to extensive search spaces, further amplified by the incorporation of somatic mutations, pathogen genomes, unannotated open reading frames, and post-translational modifications. This inflation in search space leads to an increase in random high-scoring matches, resulting in fewer identifications at a given false discovery rate. Peptide-spectrum match rescoring has emerged as a machine learning-based solution to address challenges in mass spectrometry-based immunopeptidomics data analysis. It involves post-processing unfiltered spectrum annotations to better distinguish between correct and incorrect peptide-spectrum matches. Recently, features based on predicted peptidoform properties, including fragment ion intensities, retention time, and collisional cross section, have been used to improve the accuracy and sensitivity of immunopeptide identification. In this review, we describe the diverse bioinformatics pipelines that are currently available for peptide-spectrum match rescoring and discuss how they can be used for the analysis of immunopeptidomics data. Finally, we provide insights into current and future machine learning solutions to boost immunopeptide identification.

A large-scale study of peptide features defining immunogenicity of cancer neo-epitopes

Accurate prediction of immunogenicity for neo-epitopes arising from a cancer associated mutation is a crucial step in many bioinformatics pipelines that predict outcome of checkpoint blockade treatments or that aim to design personalised cancer immunotherapies and vaccines. In this study, we performed a comprehensive analysis of peptide features relevant for prediction of immunogenicity using the Cancer Epitope Database and Analysis Resource (CEDAR), a curated database of cancer epitopes with experimentally validated immunogenicity annotations from peer-reviewed publications. The developed model, ICERFIRE (ICore-based Ensemble Random Forest for neo-epitope Immunogenicity pREdiction), extracts the predicted ICORE from the full neo-epitope as input, i.e. the nested peptide with the highest predicted major histocompatibility complex (MHC) binding potential combined with its predicted likelihood of antigen presentation (%Rank). Key additional features integrated into the model include assessment of the BLOSUM mutation score of the neo-epitope, and antigen expression levels of the wild-type counterpart which is often reflecting a neo-epitope's abundance. We demonstrate improved and robust performance of ICERFIRE over existing immunogenicity and epitope prediction models, both in cross-validation and on external validation datasets.

Challenges in developing personalized neoantigen cancer vaccines

The recent success of cancer immunotherapies has highlighted the benefit of harnessing the immune system for cancer treatment. Vaccines have a long history of promoting immunity to pathogens and, consequently, vaccines targeting cancer neoantigens have been championed as a tool to direct and amplify immune responses against tumours while sparing healthy tissue. In recent years, extensive preclinical research and more than one hundred clinical trials have tested different strategies of neoantigen discovery and vaccine formulations. However, despite the enthusiasm for neoantigen vaccines, proof of unequivocal efficacy has remained beyond reach for the majority of clinical trials. In this Review, we focus on the key obstacles pertaining to vaccine design and tumour environment that remain to be overcome in order to unleash the true potential of neoantigen vaccines in cancer therapy.

MHCpLogics: an interactive machine learning-based tool for unsupervised data visualization and cluster analysis of immunopeptidomes

The major histocompatibility complex (MHC) encodes a range of immune response genes, including the human leukocyte antigens (HLAs) in humans. These molecules bind peptide antigens and present them on the cell surface for T cell recognition. The repertoires of peptides presented by HLA molecules are termed immunopeptidomes. The highly polymorphic nature of the genres that encode the HLA molecules confers allotype-specific differences in the sequences of bound ligands. Allotype-specific ligand preferences are often defined by peptide-binding motifs. Individuals express up to six classical class I HLA allotypes, which likely present peptides displaying different binding motifs. Such complex datasets make the deconvolution of immunopeptidomic data into allotype-specific contributions and further dissection of binding-specificities challenging. Herein, we developed MHCpLogics as an interactive machine learning-based tool for mining peptide-binding sequence motifs and visualization of immunopeptidome data across complex datasets. We showcase the functionalities of MHCpLogics by analyzing both in-house and published mono- and multi-allelic immunopeptidomics data. The visualization modalities of MHCpLogics allow users to inspect clustered sequences down to individual peptide components and to examine broader sequence patterns within multiple immunopeptidome datasets. MHCpLogics can deconvolute large immunopeptidome datasets enabling the interrogation of clusters for the segregation of allotype-specific peptide sequence motifs, identification of sub-peptidome motifs, and the exportation of clustered peptide sequence lists. The tool facilitates rapid inspection of immunopeptidomes as a resource for the immunology and vaccine communities. MHCpLogics is a standalone application available via an executable installation at: https://github.com/PurcellLab/MHCpLogics.

Development and use of machine learning algorithms in vaccine target selection

Computer-aided discovery of vaccine targets has become a cornerstone of rational vaccine design. In this article, I discuss how Machine Learning (ML) can inform and guide key computational steps in rational vaccine design concerned with the identification of B and T cell epitopes and correlates of protection. I provide examples of ML models, as well as types of data and predictions for which they are built. I argue that interpretable ML has the potential to improve the identification of immunogens also as a tool for scientific discovery, by helping elucidate the molecular processes underlying vaccine-induced immune responses. I outline the limitations and challenges in terms of data availability and method development that need to be addressed to bridge the gap between advances in ML predictions and their translational application to vaccine design.

HLAEquity: Examining biases in pan-allele peptide-HLA binding predictors

Peptide-HLA (pHLA) binding prediction is essential in screening peptide candidates for personalized peptide vaccines. Machine learning (ML) pHLA binding prediction tools are trained on vast amounts of data and are effective in screening peptide candidates. Most ML models report the ability to generalize to HLA alleles unseen during training ("pan-allele" models). However, the use of datasets with imbalanced allele content raises concerns about biased model performance. First, we examine the data bias of two ML-based pan-allele pHLA binding predictors. We find that the pHLA datasets overrepresent alleles from geographic populations of high-income countries. Second, we show that the identified data bias is perpetuated within ML models, leading to algorithmic bias and subpar performance for alleles expressed in low-income geographic populations. We draw attention to the potential therapeutic consequences of this bias, and we challenge the use of the term "pan-allele" to describe models trained with currently available public datasets.

Historical perspective and future directions: computational science in immuno-oncology

Immuno-oncology holds promise for transforming patient care having achieved durable clinical response rates across a variety of advanced and metastatic cancers. Despite these achievements, only a minority of patients respond to immunotherapy, underscoring the importance of elucidating molecular mechanisms responsible for response and resistance to inform the development and selection of treatments. Breakthroughs in molecular sequencing technologies have led to the generation of an immense amount of genomic and transcriptomic sequencing data that can be mined to uncover complex tumor-immune interactions using computational tools. In this review, we discuss existing and emerging computational methods that contextualize the composition and functional state of the tumor microenvironment, infer the reactivity and clonal dynamics from reconstructed immune cell receptor repertoires, and predict the antigenic landscape for immune cell recognition. We further describe the advantage of multi-omics analyses for capturing multidimensional relationships and artificial intelligence techniques for integrating omics data with histopathological and radiological images to encapsulate patterns of treatment response and tumor-immune biology. Finally, we discuss key challenges impeding their widespread use and clinical application and conclude with future perspectives. We are hopeful that this review will both serve as a guide for prospective researchers seeking to use existing tools for scientific discoveries and inspire the optimization or development of novel tools to enhance precision, ultimately expediting advancements in immunotherapy that improve patient survival and quality of life.

Multiple instance learning to predict immune checkpoint blockade efficacy using neoantigen candidates

Previous studies showed that the neoantigen candidate load is an imperfect predictor of immune checkpoint blockade (ICB) efficacy. Further studies provided evidence that the response to ICB is also affected by the qualitative properties of a few or even single candidates, limiting the predictive power based on candidate quantity alone. Here, we predict ICB efficacy based on neoantigen candidates and their neoantigen features in the context of the mutation type, using Multiple-Instance Learning via Embedded Instance Selection (MILES). Multiple instance learning is a type of supervised machine learning that classifies labeled bags that are formed by a set of unlabeled instances. MILES performed better compared with neoantigen candidate load alone for low-abundant fusion genes in renal cell carcinoma. Our findings suggest that MILES is an appropriate method to predict the efficacy of ICB therapy based on neoantigen candidates without requiring direct T cell response information.

Machine learning methods and harmonized datasets improve immunogenic neoantigen prediction

The accurate selection of neoantigens that bind to class I human leukocyte antigen (HLA) and are recognized by autologous T cells is a crucial step in many cancer immunotherapy pipelines. We reprocessed whole-exome sequencing and RNA sequencing (RNA-seq) data from 120 cancer patients from two external large-scale neoantigen immunogenicity screening assays combined with an in-house dataset of 11 patients and identified 46,017 somatic single-nucleotide variant mutations and 1,781,445 neo-peptides, of which 212 mutations and 178 neo-peptides were immunogenic. Beyond features commonly used for neoantigen prioritization, factors such as the location of neo-peptides within protein HLA presentation hotspots, binding promiscuity, and the role of the mutated gene in oncogenicity were predictive for immunogenicity. The classifiers accurately predicted neoantigen immunogenicity across datasets and improved their ranking by up to 30%. Besides insights into machine learning methods for neoantigen ranking, we have provided homogenized datasets valuable for developing and benchmarking companion algorithms for neoantigen-based immunotherapies.

Mass Spectrometric Profiling of HLA-B44 Peptidomes Provides Evidence for Tapasin-Mediated Tryptophan Editing

The extreme polymorphisms of HLA class I proteins result in structural variations in their peptide binding sites to achieve diversity in Ag presentation. External factors could independently constrict or alter HLA class I peptide repertoires. Such effects of the assembly factor tapasin were assessed for HLA-B*44:05 (Y116) and a close variant, HLA-B*44:02 (D116), which have low and high tapasin dependence, respectively, for their cell surface expression. Analyses of the HLA-B*44:05 peptidomes in the presence and absence of tapasin reveal that peptides with C-terminal tryptophans and higher predicted affinities are preferentially selected by tapasin, coincident with reduced frequencies of peptides with other C-terminal amino acids, including leucine. Comparisons of the HLA-B*44:05 and HLA-B*44:02 peptidomes indicate the expected structure-based alterations near the peptide C termini, but also C-terminal amino acid frequency and predicted affinity changes among the unique and shared peptide groups for B*44:02 and B*44:05. Overall, these findings indicate that the presence of tapasin and the tapasin dependence of assembly alter HLA class I peptide-binding preferences at the peptide C terminus. The particular C-terminal amino acid preferences that are altered by tapasin are expected to be determined by the intrinsic peptide-binding specificities of HLA class I allotypes. Additionally, the findings suggest that tapasin deficiency and reduced tapasin dependence expand the permissive affinities of HLA class I-bound peptides, consistent with prior findings that HLA class I allotypes with low tapasin dependence have increased breadth of CD8+ T cell epitope presentation and are more protective in HIV infections.

Driver mutations in GNAQ and GNA11 genes as potential targets for precision immunotherapy in uveal melanoma patients

Uveal melanoma (UM) is the most common ocular malignancy in adults. Nearly 95% of UM patients carry the mutually exclusive mutations in the homologous genes GNAQ (amino acid change Q209L/Q209P) and GNA11 (aminoacid change Q209L). UM is located in an immunosuppressed organ and does not suffer immunoediting. Therefore, we hypothesize that driver mutations in GNAQ/11 genes could be recognized by the immune system. Genomic and transcriptomic data from primary uveal tumors were collected from the TCGA-UM dataset (n = 80) and used to assess the immunogenic potential for GNAQ/GNA11 Q209L/Q209P mutations using a variety of tools and HLA type information. All prediction tools showed stronger GNAQ/11 Q209L binding to HLA than GNAQ/11 Q209P. The immunogenicity analysis revealed that Q209L is likely to be presented by more than 73% of individuals in 1000 G databases whereas Q209P is only predicted to be presented in 24% of individuals. GNAQ/11 Q209L showed a higher likelihood to be presented by HLA-I molecules than almost all driver mutations analyzed. Finally, samples carrying Q209L had a higher immune-reactive phenotype. Regarding cancer risk, seven HLA genotypes with low Q209L affinity show higher frequency in uveal melanoma patients than in the general population. However, no clear association was found between any HLA genotype and survival. Results suggest a high potential immunogenicity of the GNAQ/11 Q209L variant that could allow the generation of novel therapeutic tools to treat UM like neoantigen vaccinations.

ProsperousPlus: a one-stop and comprehensive platform for accurate protease-specific substrate cleavage prediction and machine-learning model construction

Proteases contribute to a broad spectrum of cellular functions. Given a relatively limited amount of experimental data, developing accurate sequence-based predictors of substrate cleavage sites facilitates a better understanding of protease functions and substrate specificity. While many protease-specific predictors of substrate cleavage sites were developed, these efforts are outpaced by the growth of the protease substrate cleavage data. In particular, since data for 100+ protease types are available and this number continues to grow, it becomes impractical to publish predictors for new protease types, and instead it might be better to provide a computational platform that helps users to quickly and efficiently build predictors that address their specific needs. To this end, we conceptualized, developed, tested and released a versatile bioinformatics platform, ProsperousPlus, that empowers users, even those with no programming or little bioinformatics background, to build fast and accurate predictors of substrate cleavage sites. ProsperousPlus facilitates the use of the rapidly accumulating substrate cleavage data to train, empirically assess and deploy predictive models for user-selected substrate types. Benchmarking tests on test datasets show that our platform produces predictors that on average exceed the predictive performance of current state-of-the-art approaches. ProsperousPlus is available as a webserver and a stand-alone software package at http://prosperousplus.unimelb-biotools.cloud.edu.au/.

Neoantigen identification: Technological advances and challenges

Neoantigens have emerged as promising targets for cutting-edge immunotherapies, such as cancer vaccines and adoptive cell therapy. These neoantigens are unique to tumors and arise exclusively from somatic mutations or non-genomic aberrations in tumor proteins. They encompass a wide range of alterations, including genomic mutations, post-transcriptomic variants, and viral oncoproteins. With the advancements in technology, the identification of immunogenic neoantigens has seen rapid progress, raising new opportunities for enhancing their clinical significance. Prediction of neoantigens necessitates the acquisition of high-quality samples and sequencing data, followed by mutation calling. Subsequently, the pipeline involves integrating various tools that can predict the expression, processing, binding, and recognition potential of neoantigens. However, the continuous improvement of computational tools is constrained by the availability of datasets which contain validated immunogenic neoantigens. This review article aims to provide a comprehensive summary of the current knowledge as well as limitations in neoantigen prediction and validation. Additionally, it delves into the origin and biological role of neoantigens, offering a deeper understanding of their significance in the field of cancer immunotherapy. This article thus seeks to contribute to the ongoing efforts to harness neoantigens as powerful weapons in the fight against cancer.

TAPIR: a T-cell receptor language model for predicting rare and novel targets

T-cell receptors (TCRs) are involved in most human diseases, but linking their sequences with their targets remains an unsolved grand challenge in the field. In this study, we present TAPIR (T-cell receptor and Peptide Interaction Recognizer), a T-cell receptor (TCR) language model that predicts TCR-target interactions, with a focus on novel and rare targets. TAPIR employs deep convolutional neural network (CNN) encoders to process TCR and target sequences across flexible representations (e.g., beta-chain only, unknown MHC allele, etc.) and learns patterns of interactivity via several training tasks. This flexibility allows TAPIR to train on more than 50k either paired (alpha and beta chain) or unpaired TCRs (just alpha or beta chain) from public and proprietary databases against 1933 unique targets. TAPIR demonstrates state-of-the-art performance when predicting TCR interactivity against common benchmark targets and is the first method to demonstrate strong performance when predicting TCR interactivity against novel targets, where no examples are provided in training. TAPIR is also capable of predicting TCR interaction against MHC alleles in the absence of target information. Leveraging these capabilities, we apply TAPIR to cancer patient TCR repertoires and identify and validate a novel and potent anti-cancer T-cell receptor against a shared cancer neoantigen target (PIK3CA H1047L). We further show how TAPIR, when extended with a generative neural network, is capable of directly designing T-cell receptor sequences that interact with a target of interest.

A robust deep learning workflow to predict CD8 + T-cell epitopes

BACKGROUND

T-cells play a crucial role in the adaptive immune system by triggering responses against cancer cells and pathogens, while maintaining tolerance against self-antigens, which has sparked interest in the development of various T-cell-focused immunotherapies. However, the identification of antigens recognised by T-cells is low-throughput and laborious. To overcome some of these limitations, computational methods for predicting CD8 + T-cell epitopes have emerged. Despite recent developments, most immunogenicity algorithms struggle to learn features of peptide immunogenicity from small datasets, suffer from HLA bias and are unable to reliably predict pathology-specific CD8 + T-cell epitopes.

METHODS

We developed TRAP (T-cell recognition potential of HLA-I presented peptides), a robust deep learning workflow for predicting CD8 + T-cell epitopes from MHC-I presented pathogenic and self-peptides. TRAP uses transfer learning, deep learning architecture and MHC binding information to make context-specific predictions of CD8 + T-cell epitopes. TRAP also detects low-confidence predictions for peptides that differ significantly from those in the training datasets to abstain from making incorrect predictions. To estimate the immunogenicity of pathogenic peptides with low-confidence predictions, we further developed a novel metric, RSAT (relative similarity to autoantigens and tumour-associated antigens), as a complementary to 'dissimilarity to self' from cancer studies.

RESULTS

TRAP was used to identify epitopes from glioblastoma patients as well as SARS-CoV-2 peptides, and it outperformed other algorithms in both cancer and pathogenic settings. TRAP was especially effective at extracting immunogenicity-associated properties from restricted data of emerging pathogens and translating them onto related species, as well as minimising the loss of likely epitopes in imbalanced datasets. We also demonstrated that the novel metric termed RSAT was able to estimate immunogenic of pathogenic peptides of various lengths and species. TRAP implementation is available at: https://github.com/ChloeHJ/TRAP .

CONCLUSIONS

This study presents a novel computational workflow for accurately predicting CD8 + T-cell epitopes to foster a better understanding of antigen-specific T-cell response and the development of effective clinical therapeutics.

Prioritizing candidate peptides for cancer vaccines through predicting peptide presentation by HLA-I proteins

Cancer (treatment) vaccines that are made of neoantigens, or peptides unique to tumor cells due to somatic mutations, have emerged as a promising method to reinvigorate the immune response against cancer. A key step to prioritizing neoantigens for cancer vaccines is computationally predicting which neoantigens are presented on the cell surface by a human leukocyte antigen (HLA). We propose to address this challenge by training a neural network using mass spectrometry (MS) data composed of peptides presented by at least one of several HLAs of a subject. We embed the neural network within a mixture model and train the neural network by maximizing the likelihood of the mixture model. After evaluating our method using data sets where the peptide presentation status was known, we applied it to analyze somatic mutations of 60 melanoma patients and identified a group of neoantigens more immunogenic in tumor cells than in normal cells. Moreover, neoantigen burden estimated by our method was significantly associated with a measurement of the immune system activity, suggesting these neoantigens could induce an immune response.

The binding profile of SARS-CoV-2 with human leukocyte antigen polymorphisms reveals critical alleles involved in immune evasion

The COVID-19 pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), astonished the world and led to millions of deaths. Due to viral new mutations and immune evasion, SARS-CoV-2 ranked first in transmission and influence. The binding affinity of human leukocyte antigen (HLA) polymorphisms to SARS-CoV-2 might be related to immune escape, but the mechanisms remained unclear. In this study, we obtained the binding affinity of SARS-CoV-2 strains with different HLA proteins and identified 31 risk alleles. Subsequent structural predictions identified 10 active binding sites in these HLA proteins that may promote immune evasion. Particularly, we also found that the weak binding ability with HLA class I polymorphisms could contribute to the immune evasion of Omicron. These findings suggest important implications for preventing the immune evasion of SARS-CoV-2 and providing new insights for the vaccine design.

Antigen presentation in cancer - mechanisms and clinical implications for immunotherapy

Over the past decade, the emergence of effective immunotherapies has revolutionized the clinical management of many types of cancers. However, long-term durable tumour control is only achieved in a fraction of patients who receive these therapies. Understanding the mechanisms underlying clinical response and resistance to treatment is therefore essential to expanding the level of clinical benefit obtained from immunotherapies. In this Review, we describe the molecular mechanisms of antigen processing and presentation in tumours and their clinical consequences. We examine how various aspects of the antigen-presentation machinery (APM) shape tumour immunity. In particular, we discuss genomic variants in HLA alleles and other APM components, highlighting their influence on the immunopeptidomes of both malignant cells and immune cells. Understanding the APM, how it is regulated and how it changes in tumour cells is crucial for determining which patients will respond to immunotherapy and why some patients develop resistance. We focus on recently discovered molecular and genomic alterations that drive the clinical outcomes of patients receiving immune-checkpoint inhibitors. An improved understanding of how these variables mediate tumour-immune interactions is expected to guide the more precise administration of immunotherapies and reveal potentially promising directions for the development of new immunotherapeutic approaches.

Improved prediction of MHC-peptide binding using protein language models

Major histocompatibility complex Class I (MHC-I) molecules bind to peptides derived from intracellular antigens and present them on the surface of cells, allowing the immune system (T cells) to detect them. Elucidating the process of this presentation is essential for regulation and potential manipulation of the cellular immune system. Predicting whether a given peptide binds to an MHC molecule is an important step in the above process and has motivated the introduction of many computational approaches to address this problem. NetMHCPan, a pan-specific model for predicting binding of peptides to any MHC molecule, is one of the most widely used methods which focuses on solving this binary classification problem using shallow neural networks. The recent successful results of Deep Learning (DL) methods, especially Natural Language Processing (NLP-based) pretrained models in various applications, including protein structure determination, motivated us to explore their use in this problem. Specifically, we consider the application of deep learning models pretrained on large datasets of protein sequences to predict MHC Class I-peptide binding. Using the standard performance metrics in this area, and the same training and test sets, we show that our models outperform NetMHCpan4.1, currently considered as the-state-of-the-art.

Targeting alternative splicing in cancer immunotherapy

Tumor immunotherapy has made great progress in cancer treatment but still faces several challenges, such as a limited number of targetable antigens and varying responses among patients. Alternative splicing (AS) is an essential process for the maturation of nearly all mammalian mRNAs. Recent studies show that AS contributes to expanding cancer-specific antigens and modulating immunogenicity, making it a promising solution to the above challenges. The organoid technology preserves the individual immune microenvironment and reduces the time/economic costs of the experiment model, facilitating the development of splicing-based immunotherapy. Here, we summarize three critical roles of AS in immunotherapy: resources for generating neoantigens, targets for immune-therapeutic modulation, and biomarkers to guide immunotherapy options. Subsequently, we highlight the benefits of adopting organoids to develop AS-based immunotherapies. Finally, we discuss the current challenges in studying AS-based immunotherapy in terms of existing bioinformatics algorithms and biological technologies.

Deep neural networks predict class I major histocompatibility complex epitope presentation and transfer learn neoepitope immunogenicity

Identifying neoepitopes that elicit an adaptive immune response is a major bottleneck to developing personalized cancer vaccines. Experimental validation of candidate neoepitopes is extremely resource intensive and the vast majority of candidates are non-immunogenic, creating a needle-in-a-haystack problem. Here we address this challenge, presenting computational methods for predicting class I major histocompatibility complex (MHC-I) epitopes and identifying immunogenic neoepitopes with improved precision. The BigMHC method comprises an ensemble of seven pan-allelic deep neural networks trained on peptide-MHC eluted ligand data from mass spectrometry assays and transfer learned on data from assays of antigen-specific immune response. Compared with four state-of-the-art classifiers, BigMHC significantly improves the prediction of epitope presentation on a test set of 45,409 MHC ligands among 900,592 random negatives (area under the receiver operating characteristic = 0.9733; area under the precision-recall curve = 0.8779). After transfer learning on immunogenicity data, BigMHC yields significantly higher precision than seven state-of-the-art models in identifying immunogenic neoepitopes, making BigMHC effective in clinical settings.

A microfluidics-enabled automated workflow of sample preparation for MS-based immunopeptidomics

Mass spectrometry (MS)-based immunopeptidomics is an attractive antigen discovery method with growing clinical implications. However, the current experimental approach to extract HLA-restricted peptides requires a bulky sample source, which remains a challenge for obtaining clinical specimens. We present an innovative workflow that requires a low sample volume, which streamlines the immunoaffinity purification (IP) and C18 peptide cleanup on a single microfluidics platform with automated liquid handling and minimal sample transfers, resulting in higher assay sensitivity. We also demonstrate how the state-of-the-art data-independent acquisition (DIA) method further enhances the depth of tandem MS spectra-based peptide sequencing. Consequently, over 4,000 and 5,000 HLA-I-restricted peptides were identified from as few as 0.2 million RA957 cells and a melanoma tissue of merely 5 mg, respectively. We also identified multiple immunogenic tumor-associated antigens and hundreds of peptides derived from non-canonical protein sources. This workflow represents a powerful tool for identifying the immunopeptidome of sparse samples.

Neoantigen-specific CD8 T cells with high structural avidity preferentially reside in and eliminate tumors

The success of cancer immunotherapy depends in part on the strength of antigen recognition by T cells. Here, we characterize the T cell receptor (TCR) functional (antigen sensitivity) and structural (monomeric pMHC-TCR off-rates) avidities of 371 CD8 T cell clones specific for neoantigens, tumor-associated antigens (TAAs) or viral antigens isolated from tumors or blood of patients and healthy donors. T cells from tumors exhibit stronger functional and structural avidity than their blood counterparts. Relative to TAA, neoantigen-specific T cells are of higher structural avidity and, consistently, are preferentially detected in tumors. Effective tumor infiltration in mice models is associated with high structural avidity and CXCR3 expression. Based on TCR biophysicochemical properties, we derive and apply an in silico model predicting TCR structural avidity and validate the enrichment in high avidity T cells in patients' tumors. These observations indicate a direct relationship between neoantigen recognition, T cell functionality and tumor infiltration. These results delineate a rational approach to identify potent T cells for personalized cancer immunotherapy.

The Immunopeptidome from a Genomic Perspective: Establishing the Noncanonical Landscape of MHC Class I-Associated Peptides

Tumor antigens can emerge through multiple mechanisms, including translation of noncoding genomic regions. This noncanonical category of tumor antigens has recently gained attention; however, our understanding of how they recur within and between cancer types is still in its infancy. Therefore, we developed a proteogenomic pipeline based on deep learning de novo mass spectrometry (MS) to enable the discovery of noncanonical MHC class I-associated peptides (ncMAP) from noncoding regions. Considering that the emergence of tumor antigens can also involve posttranslational modifications (PTM), we included an open search component in our pipeline. Leveraging the wealth of MS-based immunopeptidomics, we analyzed data from 26 MHC class I immunopeptidomic studies across 11 different cancer types. We validated the de novo identified ncMAPs, along with the most abundant PTMs, using spectral matching and controlled their FDR to 1%. The noncanonical presentation appeared to be 5 times enriched for the A03 HLA supertype, with a projected population coverage of 55%. The data reveal an atlas of 8,601 ncMAPs with varying levels of cancer selectivity and suggest 17 cancer-selective ncMAPs as attractive therapeutic targets according to a stringent cutoff. In summary, the combination of the open-source pipeline and the atlas of ncMAPs reported herein could facilitate the identification and screening of ncMAPs as targets for T-cell therapies or vaccine development.

Highly Networked SARS-CoV-2 Peptides Elicit T Cell Responses with Enhanced Specificity

Identifying SARS-CoV-2-specific T cell epitope-derived peptides is critical for the development of effective vaccines and measuring the duration of specific SARS-CoV-2 cellular immunity. In this regard, we previously identified T cell epitope-derived peptides within topologically and structurally essential regions of SARS-CoV-2 spike and nucleocapsid proteins by applying an immunoinformatics pipeline. In this study, we selected 30 spike- and nucleocapsid-derived peptides and assessed whether these peptides induce T cell responses and avoid major mutations found in SARS-CoV-2 variants of concern. Our peptide pool was highly specific, with only a single peptide driving cross-reactivity in people unexposed to SARS-COV-2, and immunogenic, inducing a polyfunctional response in CD4+ and CD8+ T cells from COVID-19 recovered individuals. All peptides were immunogenic and individuals recognized broad and diverse peptide repertoires. Moreover, our peptides avoided most mutations/deletions associated with all four SARS-CoV-2 variants of concern while retaining their physicochemical properties even when genetic changes are introduced. This study contributes to an evolving definition of individual CD4+ and CD8+ T cell epitopes that can be used for specific diagnostic tools for SARS-CoV-2 T cell responses and is relevant to the development of variant-resistant and durable T cell-stimulating vaccines.

The biogenesis of the immunopeptidome

The immunopeptidome is the repertoire of peptides bound and presented by the MHC class I, class II, and non-classical molecules. The peptides are produced by the degradation of most cellular proteins, and in some cases, peptides are produced from extracellular proteins taken up by the cells. This review attempts to first describe some of its known and well-accepted concepts, and next, raise some questions about a few of the established dogmas in this field: The production of novel peptides by splicing is questioned, suggesting here that spliced peptides are extremely rare, if existent at all. The degree of the contribution to the immunopeptidome by degradation of cellular protein by the proteasome is doubted, therefore this review attempts to explain why it is likely that this contribution to the immunopeptidome is possibly overstated. The contribution of defective ribosome products (DRiPs) and non-canonical peptides to the immunopeptidome is noted and methods are suggested to quantify them. In addition, the common misconception that the MHC class II peptidome is mostly derived from extracellular proteins is noted, and corrected. It is stressed that the confirmation of sequence assignments of non-canonical and spliced peptides should rely on targeted mass spectrometry using spiking-in of heavy isotope-labeled peptides. Finally, the new methodologies and modern instrumentation currently available for high throughput kinetics and quantitative immunopeptidomics are described. These advanced methods open up new possibilities for utilizing the big data generated and taking a fresh look at the established dogmas and reevaluating them critically.

The immunopeptidome landscape associated with T cell infiltration, inflammation and immune editing in lung cancer

One key barrier to improving efficacy of personalized cancer immunotherapies that are dependent on the tumor antigenic landscape remains patient stratification. Although patients with CD3⁺CD8⁺ T cell-inflamed tumors typically show better response to immune checkpoint inhibitors, it is still unknown whether the immunopeptidome repertoire presented in highly inflamed and noninflamed tumors is substantially different. We surveyed 61 tumor regions and adjacent nonmalignant lung tissues from 8 patients with lung cancer and performed deep antigen discovery combining immunopeptidomics, genomics, bulk and spatial transcriptomics, and explored the heterogeneous expression and presentation of tumor (neo)antigens. In the present study, we associated diverse immune cell populations with the immunopeptidome and found a relatively higher frequency of predicted neoantigens located within HLA-I presentation hotspots in CD3⁺CD8⁺ T cell-excluded tumors. We associated such neoantigens with immune recognition, supporting their involvement in immune editing. This could have implications for the choice of combination therapies tailored to the patient's mutanome and immune microenvironment.

STMHCpan, an accurate Star-Transformer-based extensible framework for predicting MHC I allele binding peptides

Peptide-major histocompatibility complex I (MHC I) binding affinity prediction is crucial for vaccine development, but existing methods face limitations such as small datasets, model overfitting due to excessive parameters and suboptimal performance. Here, we present STMHCPan (STAR-MHCPan), an open-source package based on the Star-Transformer model, for MHC I binding peptide prediction. Our approach introduces an attention mechanism to improve the deep learning network architecture and performance in antigen prediction. Compared with classical deep learning algorithms, STMHCPan exhibits improved performance with fewer parameters in receptor affinity training. Furthermore, STMHCPan outperforms existing ligand benchmark datasets identified by mass spectrometry. It can also handle peptides of arbitrary length and is highly scalable for predicting T-cell responses. Our software is freely available for use, training and extension through Github (https://github.com/Luckysoutheast/STMHCPan.git).

Adoptive neoantigen-reactive T cell therapy: improvement strategies and current clinical researches

Neoantigens generated by non-synonymous mutations of tumor genes can induce activation of neoantigen-reactive T (NRT) cells which have the ability to resist the growth of tumors expressing specific neoantigens. Immunotherapy based on NRT cells has made preeminent achievements in melanoma and other solid tumors. The process of manufacturing NRT cells includes identification of neoantigens, preparation of neoantigen expression vectors or peptides, induction and activation of NRT cells, and analysis of functions and phenotypes. Numerous improvement strategies have been proposed to enhance the potency of NRT cells by engineering TCR, promoting infiltration of T cells and overcoming immunosuppressive factors in the tumor microenvironment. In this review, we outline the improvement of the preparation and the function assessment of NRT cells, and discuss the current status of clinical trials related to NRT cell immunotherapy.