Critical review of conformational B-cell epitope prediction methods

Molecular characterization, B-cell linear epitopes identification and key amino acids selection of the sesame allergen Ses i 5

Ses i 5 is an important sesame allergen and no studies have been reported on the epitope identification of this allergen. Epitopes are the antigenic determinants on the surface of allergens, which are the basis of intensive study on protein antigenicity. In this paper, we analyzed the secondary structure, tertiary structure, hydrophilicity, antigenic index, flexibility, surface accessibility, and homology of the Ses i 5. In addition, we designed and synthesized overlapping peptides covering the entire amino acid sequence of Ses i 5, each overlapping peptide contains 15 amino acids with 5 amino acid repeats. The IgE binding capacity of each peptide was assessed by immune slot blot microarray. Ultimately, we identified seven B-cell linear epitopes of Ses i 5. Furthermore, the key amino acids of Ses i 5 were predicted and mutated to alanine, the changes in the IgE-binding capacity of the mutated peptides were assessed by indirect competition-ELISA (ic-ELISA). At last glycine 104 was identified as the key amino acid for IgE binding in Ses i 5. These studies will contribute to the deep study of molecular characterization and more information on epitopes of Ses i 5, which will benefit for the further understanding of the sesame allergen.

Designing a multi-epitope vaccine against Pseudomonas aeruginosa via integrating reverse vaccinology with immunoinformatics approaches

Pseudomonas aeruginosa is a typically opportunistic pathogen responsible for a wide range of nosocomial infections. In this study, we designed two multi-epitope vaccines targeting P. aeruginosa proteins, incorporating cytotoxic T lymphocyte (CTL), helper T lymphocyte (HTL), and linear B lymphocyte (LBL) epitopes identified using reverse vaccinology and immunoinformatics approaches. The vaccines exhibited favorable physicochemical properties, including stability, solubility, and optimal molecular weight, suggesting their potential as viable candidates for vaccine development. Molecular docking studies revealed strong binding affinity to Toll-like receptors 1 (TLR1) and 2 (TLR2). Furthermore, molecular dynamics simulations confirmed the stability of the vaccine-TLR complexes over time. Immune simulation analyses indicated that the vaccines could induce robust humoral and cellular immune responses, providing a promising new approach for combating P. aeruginosa infections, particularly in the face of increasing antibiotic resistance.

A rapid approach for linear epitope vaccine profiling reveals unexpected epitope tag immunogenicity

Antibody epitope profiling is essential for assessing the robustness of vaccine-induced immune responses, particularly while in development. Despite advancements in computational tools, high throughput experimental epitope validation remains an important step. Here, we describe a readily accessible method for rapid linear epitope profiling using phage-displayed oligo pools in combination with Nanopore deep sequencing. We applied this approach to TeeVax3, a Group A Streptococcus vaccine candidate, to investigate the antibody response generated in a pre-clinical rabbit model and assess antigen immunogenicity. Surprisingly, we found a strong bias in antibody binding response towards the N-terminal epitope tag used for purification. These tags are widely reported to have low immunogenicity and are frequently left uncleaved in pre-clinical studies. We further confirmed that the observed immune response against the epitope tag dominated even the conformational binding response and, using synthetic peptides, narrowed the epitope down to a set of 10 residues inclusive of the Histidine residues. Our findings highlight the importance of epitope-tag removal in pre-clinical studies and demonstrate the utility of rapid Nanopore sequencing for early-stage vaccine evaluation.

Advancing vaccine development in genomic era: a paradigm shift in vaccine discovery

The issue of antibiotic resistance is increasing with time because of the quick rise of microbial strains. Overuse of antibiotics has led to multidrug-resistant, pan-drug-resistant, and extensively drug-resistant bacterial strains, which have worsened the situation. Different techniques have been considered and applied to combat this issue, such as developing new antibiotics, practicing antibiotic stewardship, improving hygiene levels, and controlling antibiotic overuse. Vaccine development made a substantial contribution to overcoming this issue, although it has been underestimated. In the recent era, reverse vaccinology has contributed to developing different kinds of vaccines against pathogens, revolutionizing the vaccine development process. Reverse vaccinology helps to prioritize better vaccine candidates by using various tools to filter the pathogen's complete genome. In this review, we will shed light on computational vaccine designing, immunoinformatic tools, genomic and proteomic data, and the challenges and success stories of computational vaccine designing.

Compromising the immunogenicity of diphtheria toxin-based immunotoxins through epitope engineering: An in silico approach

Immunotoxins are genetically engineered recombinant proteins consisting of a targeting moiety, such as an antibody, and a cytotoxic toxin moiety of microbial origin. Pseudomonas exotoxin A and diphtheria toxin (DT) have been abundantly used in immunotoxins, with the latter applied as the toxin moiety of the FDA-approved drug Denileukin diftitox (ONTAK®). However, the use of immunotoxins provokes an adverse immune response in the host body against the toxin moiety, limiting their efficacy. In silico approaches have received increasing attention in protein engineering. In this study, the epitopes responsible for immunogenicity were identified through multiple platforms. By subtracting conserved and ligand-binding residues, K33, T111, and E112 were identified as common epitopes across all platforms. Substitution analysis evaluated alternative residues regarding their impact on protein stability, considering 19 different amino acid substitutions. Among the mutants explored, the T111A-E112G mutant exhibited the most destabilizing substitution for DT, thereby reducing immunogenicity. Finally, a 3D model of the mutant was generated and verified. The model was then docked with its native ligand NADH, and the complex's molecular behavior was simulated using molecular dynamics.

A Bibliometric Analysis on Multi-epitope Vaccine Development Against SARS-CoV-2: Current Status, Development, and Future Directions

The etiological agent for the coronavirus disease 2019 (COVID-19), the SARS-CoV-2, caused a global pandemic. Although mRNA, viral-vectored, DNA, and recombinant protein vaccine candidates were effective against the SARS-CoV-2 Wuhan strain, the emergence of SARS-CoV-2 variants of concern (VOCs) reduced the protective efficacies of these vaccines. This necessitates the need for effective and accelerated vaccine development against mutated VOCs. The development of multi-epitope vaccines against SARS-CoV-2 based on in silico identification of highly conserved and immunogenic epitopes is a promising strategy for future SARS-CoV-2 vaccine development. Considering the evolving landscape of the COVID-19 pandemic, we have conducted a bibliometric analysis to consolidate current findings and research trends in multi-epitope vaccine development to provide insights for future vaccine development strategies. Analysis of 102 publications on multi-epitope vaccine development against SARS-CoV-2 revealed significant growth and global collaboration, with India leading in the number of publications, along with an identification of the most prolific authors. Key journals included the Journal of Biomolecular Structure and Dynamics, while top collaborations involved Pakistan-China and India-USA. Keyword analysis showed a prominent focus on immunoinformatics, epitope prediction, and spike glycoprotein. Advances in immunoinformatics, including AI-driven epitope prediction, offer promising avenues for the development of safe and effective multi-epitope vaccines. Immunogenicity may be further improved through nanoparticle-based systems or the use of adjuvants along with real-time genomic surveillance to tailor vaccines against emerging variants.

Epitope mapping via in vitro deep mutational scanning methods and its applications

Epitope mapping is a technique employed to define the region of an antigen that elicits an immune response, providing crucial insight into the structural architecture of the antigen as well as epitope-paratope interactions. With this breadth of knowledge, immunotherapies, diagnostics, and vaccines are being developed with a rational and data-supported design. Traditional epitope mapping methods are laborious, time-intensive, and often lack the ability to screen proteins in a high-throughput manner or provide high resolution. Deep mutational scanning (DMS), however, is revolutionizing the field as it can screen all possible single amino acid mutations and provide an efficient and high-throughput way to infer the structures of both linear and three-dimensional epitopes with high resolution. Currently, more than 50 publications take this approach to efficiently identify enhancing or escaping mutations, with many then employing this information to rapidly develop broadly neutralizing antibodies, T-cell immunotherapies, vaccine platforms, or diagnostics. We provide a comprehensive review of the approaches to accomplish epitope mapping while also providing a summation of the development of DMS technology and its impactful applications.

Isolation of porcine epidemic diarrhea virus strain CHCQ-2023 from Chongqing Province and analysis of S gene recombination

BACKGROUND

In recent years, the prevalence and incidence of porcine epidemic diarrhea virus (PEDV) infection have been on the rise. The occurrence of multiviral infections and recombination mutations has led to accelerated viral evolution and reduced vaccine efficacy. In the present study, a PEDV strain was isolated from a pig farm (Chongqing Province, China) with an outbreak of porcine diarrhea, and its S gene was found to be recombinant.

RESULTS

The optimal trypsin concentration for blind passage of PEDV in Vero cells was determined to be 7.5 µg/mL. Following two blind passages of the virus in Vero cells, the virus was unable to adapt to the cells. Therefore, PEDV was blindly passaged in IPEC-J2 cells using the optimal concentration of trypsin (5 µg/mL). Next, a series of characterization experiments were performed. Recombination analyses of the isolates using software revealed that the S gene of strain CHCQ-2023 was derived from the primary parent strain PEDV-1 C and secondary parent strain SQ2014, with recombination occurring at a 3152 bp breakpoint. Furthermore, a specific B-cell antigenic epitope was predicted on the S2 subunit of the S protein.

CONCLUSION

A PEDV strain was isolated and characterized, and its S gene was characterized. The findings provide a bioinformatic basis for the study of PEDV strain variation due to genetic recombination.

A structure-based in silico analysis of the Kell blood group system

Kell is one of the most complex blood group systems, with a highly polymorphic genetic background. Extensive allelic variations in the KEL gene affect the encoded erythrocyte surface protein Kell. Genetic variants causing aberrant splicing, premature termination of protein translation, or specific amino acid exchanges lead to a variety of different phenotypes with altered Kell expression levels or changes in the antigenic properties of the Kell protein. Using an in silico structural model of the Kell protein, we analyzed the biophysical and structural context of all full-length Kell variants of known phenotype. The results provided insights regarding the 3D co-localization of antigenic Kell variants and led us to suggest several conformational epitopes on the Kell protein surface. We found a number of correlations between the properties of individual genetic variants in the Kell protein and their respective serological phenotypes, which we used as a search filter to predict potentially new immunogenic Kell variants from an in-house whole exome sequencing dataset of 19,772 exomes. Our analysis workflow and results aid blood group serologists in predicting whether a newly identified Kell genetic variant may result in a specific phenotype.

Deep learning of antibody epitopes using positional permutation vectors

Background

The accurate computational prediction of B cell epitopes can vastly reduce the cost and time required for identifying potential epitope candidates for the design of vaccines and immunodiagnostics. However, current computational tools for B cell epitope prediction perform poorly and are not fit-for-purpose, and there remains enormous room for improvement and the need for superior prediction strategies.

Results

Here we propose a novel approach that improves B cell epitope prediction by encoding epitopes as binary positional permutation vectors that represent the position and structural properties of the amino acids within a protein antigen sequence that interact with an antibody. This approach supersedes the traditional method of defining epitopes as scores per amino acid on a protein sequence, where each score reflects each amino acids predicted probability of partaking in a B cell epitope antibody interaction. In addition to defining epitopes as binary positional permutation vectors, the approach also uses the 3D macrostructure features of the unbound protein structures, and in turn uses these features to train another deep learning model on the corresponding antibody-bound protein 3D structures. This enables the algorithm to learn the key structural and physiochemical features of the unbound protein and embedded epitope that initiate the antibody binding process helping to eliminate "induced fit" biases in the training data. We demonstrate that the strategy predicts B cell epitopes with improved accuracy compared to the existing tools. Additionally, we show that this approach reliably identifies the majority of experimentally verified epitopes on the spike protein of SARS-CoV-2 not seen by the model during training and generalizes in a very robust manner on dissimilar data not seen by the model during training.

Conclusions

With the approach described herein, a primary protein sequence and a query positional permutation vector encoding a putative epitope is sufficient to predict B cell epitopes in a reliable manner, potentially advancing the use of computational prediction of B cell epitopes in biomedical research applications.

Immunogenicity risk assessment and mitigation for engineered antibody and protein therapeutics

Remarkable progress has been made in recent decades in engineering antibodies and other protein therapeutics, including enhancements to existing functions as well as the advent of novel molecules that confer biological activities previously unknown in nature. These protein therapeutics have brought major benefits to patients across multiple areas of medicine. One major ongoing challenge is that protein therapeutics can elicit unwanted immune responses (immunogenicity) in treated patients, including the generation of anti-drug antibodies. In rare and unpredictable cases, anti-drug antibodies can seriously compromise therapeutic safety and/or efficacy. Systematic deconvolution of this immunogenicity problem is confounded by the complexity of its many contributing factors and the inherent limitations of available experimental and computational methods. Nevertheless, continued progress with the assessment and mitigation of immunogenicity risk at the preclinical stage has the potential to reduce the incidence and severity of clinical immunogenicity events. This Review focuses on identifying key unsolved anti-drug antibody-related challenges and offers some pragmatic approaches towards addressing them. Examples are drawn mainly from antibodies, given that the majority of available clinical data are from this class of protein therapeutics. Plausible and seemingly tractable solutions are in sight for some immunogenicity problems, whereas other challenges will likely require completely new approaches.

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.

B cell epitope prediction by capturing spatial clustering property of the epitopes using graph attention network

Knowledge of B cell epitopes is critical to vaccine design, diagnostics, and therapeutics. As experimental validation for epitopes is time-consuming and costly, many in silico tools have been developed to computationally predict the B cell epitopes. While most methods show poor performance, deep learning methods in recent years have shown promising results. We developed a method called EpiGraph that outperformed previous methods, including those that showed a significant improvement in performance in recent years. Our model's performance can be attributed to the following factors: (1) a combination of structure and sequence feature embeddings obtained from pretrained ESM-IF1 and ESM-2 models could capture the structural and evolutionary features of B cell epitopes, (2) a graph attention network could learn the spatial proximity of B cell epitopes with high graph homophily, and (3) residual connections in the model framework mitigate the over-smoothing problem in the graph neural network. Our model achieved the highest performance on an independent benchmark dataset. The results were also consistent on a different dataset. The datasets and source codes are available at https://github.com/sj584/EpiGraph . A user-friendly web server is freely available at http://epigraph.kaist.ac.kr .

An explainable language model for antibody specificity prediction using curated influenza hemagglutinin antibodies

Despite decades of antibody research, it remains challenging to predict the specificity of an antibody solely based on its sequence. Two major obstacles are the lack of appropriate models and the inaccessibility of datasets for model training. In this study, we curated >5,000 influenza hemagglutinin (HA) antibodies by mining research publications and patents, which revealed many distinct sequence features between antibodies to HA head and stem domains. We then leveraged this dataset to develop a lightweight memory B cell language model (mBLM) for sequence-based antibody specificity prediction. Model explainability analysis showed that mBLM could identify key sequence features of HA stem antibodies. Additionally, by applying mBLM to HA antibodies with unknown epitopes, we discovered and experimentally validated many HA stem antibodies. Overall, this study not only advances our molecular understanding of the antibody response to the influenza virus but also provides a valuable resource for applying deep learning to antibody research.

Machine learning tools used for mapping some immunogenic epitopes within the major structural proteins of the bovine coronavirus (BCoV) and for the in silico design of the multiepitope-based vaccines

Introduction

BCoV is one of the significant causes of enteritis in young calves; it may also be responsible for many respiratory outbreaks in young calves. BCoV participates in the development of bovine respiratory disease complex in association with other bacterial pathogens. Our study aimed (1) to map the immunogenic epitopes (B and T cells) within the major BCoV structural proteins. These epitopes are believed to induce a robust immune response through the interaction with major histocompatibility complex (MHC class II) molecules (2) to design some novel BCoV multiepitope-based vaccines.

Materials and Methods

The goal is achieved through several integrated in silico prediction computational tools to map these epitopes within the major BCoV structural proteins. The final vaccine was constructed in conjugation with the Choleratoxin B toxin as an adjuvant. The tertiary structure of each vaccine construct was modeled through the AlphaFold2 tools. The constructed vaccine was linked to some immunostimulants such as Toll-like receptors (TLR2 and TLR4). We also predicted the affinity binding of these vaccines with this targeted protein using molecular docking. The stability and purity of each vaccine construct were assessed using the Ramachandran plot and the Z-score values. We created the in silico cloning vaccine constructs using various expression vectors through vector builder and Snap gene.

Results and discussion

The average range of major BCoV structural proteins was detected within the range of 0.4 to 0.5, which confirmed their antigen and allergic properties. The binding energy values were detected between -7.9 and -9.4 eV and also confirmed their best interaction between our vaccine construct and Toll-like receptors. Our in silico cloning method expedited the creation of vaccine constructs and established a strong basis for upcoming clinical trials and experimental validations.

Conclusion

Our designed multiepitope vaccine candidates per each BCoV structural protein showed high antigenicity, immunogenicity, non-allergic, non-toxic, and high-water solubility. Further studies are highly encouraged to validate the efficacy of these novel BCoV vaccines in the natural host.

Computational prediction of multiple antigen epitopes

MOTIVATION

Identifying antigen epitopes is essential in medical applications, such as immunodiagnostic reagent discovery, vaccine design, and drug development. Computational approaches can complement low-throughput, time-consuming, and costly experimental determination of epitopes. Currently available prediction methods, however, have moderate success predicting epitopes, which limits their applicability. Epitope prediction is further complicated by the fact that multiple epitopes may be located on the same antigen and complete experimental data is often unavailable.

RESULTS

Here, we introduce the antigen epitope prediction program ISPIPab that combines information from two feature-based methods and a docking-based method. We demonstrate that ISPIPab outperforms each of its individual classifiers as well as other state-of-the-art methods, including those designed specifically for epitope prediction. By combining the prediction algorithm with hierarchical clustering, we show that we can effectively capture epitopes that align with available experimental data while also revealing additional novel targets for future experimental investigations.

Machine-learning-based structural analysis of interactions between antibodies and antigens

Computational analysis of paratope-epitope interactions between antibodies and their corresponding antigens can facilitate our understanding of the molecular mechanism underlying humoral immunity and boost the design of new therapeutics for many diseases. The recent breakthrough in artificial intelligence has made it possible to predict protein-protein interactions and model their structures. Unfortunately, detecting antigen-binding sites associated with a specific antibody is still a challenging problem. To tackle this challenge, we implemented a deep learning model to characterize interaction patterns between antibodies and their corresponding antigens. With high accuracy, our model can distinguish between antibody-antigen complexes and other types of protein-protein complexes. More intriguingly, we can identify antigens from other common protein binding regions with an accuracy of higher than 70% even if we only have the epitope information. This indicates that antigens have distinct features on their surface that antibodies can recognize. Additionally, our model was unable to predict the partnerships between antibodies and their particular antigens. This result suggests that one antigen may be targeted by more than one antibody and that antibodies may bind to previously unidentified proteins. Taken together, our results support the precision of antibody-antigen interactions while also suggesting positive future progress in the prediction of specific pairing.

Predicting monoclonal antibody binding sequences from a sparse sampling of all possible sequences

Previous work has shown that binding of target proteins to a sparse, unbiased sample of all possible peptide sequences is sufficient to train a machine learning model that can then predict, with statistically high accuracy, target binding to any possible peptide sequence of similar length. Here, highly sequence-specific molecular recognition is explored by measuring binding of 8 monoclonal antibodies (mAbs) with specific linear cognate epitopes to an array containing 121,715 near-random sequences about 10 residues in length. Network models trained on resulting sequence-binding values are used to predict the binding of each mAb to its cognate sequence and to an in silico generated one million random sequences. The model always ranks the binding of the cognate sequence in the top 100 sequences, and for 6 of the 8 mAbs, the cognate sequence ranks in the top ten. Practically, this approach has potential utility in selecting highly specific mAbs for therapeutics or diagnostics. More fundamentally, this demonstrates that very sparse random sampling of a large amino acid sequence spaces is sufficient to generate comprehensive models predictive of highly specific molecular recognition.

Computational Methods to Predict Conformational B-Cell Epitopes

Accurate computational prediction of B-cell epitopes can greatly enhance biomedical research and rapidly advance efforts to develop therapeutics, monoclonal antibodies, vaccines, and immunodiagnostic reagents. Previous research efforts have primarily focused on the development of computational methods to predict linear epitopes rather than conformational epitopes; however, the latter is much more biologically predominant. Several conformational B-cell epitope prediction methods have recently been published, but their predictive performances are weak. Here, we present a review of the latest computational methods and assess their performances on a diverse test set of 29 non-redundant unbound antigen structures. Our results demonstrate that ISPIPab performs better than most methods and compares favorably with other recent antigen-specific methods. Finally, we suggest new strategies and opportunities to improve computational predictions of conformational B-cell epitopes.

Computational Prediction of Multiple Antigen Epitopes

Motivation

Identifying antigen epitopes is essential in medical applications, such as immunodiagnostic reagent discovery, vaccine design, and drug development. Computational approaches can complement low-throughput, time-consuming, and costly experimental determination of epitopes. Currently available prediction methods, however, have moderate success predicting epitopes, which limits their applicability. Epitope prediction is further complicated by the fact that multiple epitopes may be located on the same antigen and complete experimental data is often unavailable.

Results

Here, we introduce the antigen epitope prediction program ISPIPab that combines information from two feature-based methods and a docking-based method. We demonstrate that ISPIPab outperforms each of its individual classifiers as well as other state-of-the-art methods, including those designed specifically for epitope prediction. By combining the prediction algorithm with hierarchical clustering, we show that we can effectively capture epitopes that align with available experimental data while also revealing additional novel targets for future experimental investigations.

Contact

raji@yu.edu.

Supplementary information

Supplementary data are available at Bioinformatics online.

Molecular Modeling Methods in the Development of Affine and Specific Protein-Binding Agents

High-affinity and specific agents are widely applied in various areas, including diagnostics, scientific research, and disease therapy (as drugs and drug delivery systems). It takes significant time to develop them. For this reason, development of high-affinity agents extensively utilizes computer methods at various stages for the analysis and modeling of these molecules. The review describes the main affinity and specific agents, such as monoclonal antibodies and their fragments, antibody mimetics, aptamers, and molecularly imprinted polymers. The methods of their obtaining as well as their main advantages and disadvantages are briefly described, with special attention focused on the molecular modeling methods used for their analysis and development.

Immunodominant extracellular loops of Treponema pallidum FadL outer membrane proteins elicit antibodies with opsonic and growth-inhibitory activities

The global resurgence of syphilis has created a potent stimulus for vaccine development. To identify potentially protective antibodies (Abs) against Treponema pallidum (TPA), we used Pyrococcus furiosus thioredoxin (PfTrx) to display extracellular loops (ECLs) from three TPA outer membrane protein families (outer membrane factors for efflux pumps, eight-stranded β-barrels, and FadLs) to assess their reactivity with immune rabbit serum (IRS). Five ECLs from the FadL orthologs TP0856, TP0858 and TP0865 were immunodominant. Rabbits and mice immunized with these five PfTrx constructs produced ECL-specific Abs that promoted opsonophagocytosis of TPA by rabbit peritoneal and murine bone marrow-derived macrophages at levels comparable to IRS and mouse syphilitic serum. ECL-specific rabbit and mouse Abs also impaired viability, motility, and cellular attachment of spirochetes during in vitro cultivation. The results support the use of ECL-based vaccines and suggest that ECL-specific Abs promote spirochete clearance via Fc receptor-independent as well as Fc receptor-dependent mechanisms.

Geometric epitope and paratope prediction

MOTIVATION

Identifying the binding sites of antibodies is essential for developing vaccines and synthetic antibodies. In this article, we investigate the optimal representation for predicting the binding sites in the two molecules and emphasize the importance of geometric information.

RESULTS

Specifically, we compare different geometric deep learning methods applied to proteins' inner (I-GEP) and outer (O-GEP) structures. We incorporate 3D coordinates and spectral geometric descriptors as input features to fully leverage the geometric information. Our research suggests that different geometrical representation information is useful for different tasks. Surface-based models are more efficient in predicting the binding of the epitope, while graph models are better in paratope prediction, both achieving significant performance improvements. Moreover, we analyze the impact of structural changes in antibodies and antigens resulting from conformational rearrangements or reconstruction errors. Through this investigation, we showcase the robustness of geometric deep learning methods and spectral geometric descriptors to such perturbations.

AVAILABILITY AND IMPLEMENTATION

The python code for the models, together with the data and the processing pipeline, is open-source and available at https://github.com/Marco-Peg/GEP.

Prediction of Paratope-Epitope Pairs Using Convolutional Neural Networks

Antibodies play a central role in the adaptive immune response of vertebrates through the specific recognition of exogenous or endogenous antigens. The rational design of antibodies has a wide range of biotechnological and medical applications, such as in disease diagnosis and treatment. However, there are currently no reliable methods for predicting the antibodies that recognize a specific antigen region (or epitope) and, conversely, epitopes that recognize the binding region of a given antibody (or paratope). To fill this gap, we developed ImaPEp, a machine learning-based tool for predicting the binding probability of paratope-epitope pairs, where the epitope and paratope patches were simplified into interacting two-dimensional patches, which were colored according to the values of selected features, and pixelated. The specific recognition of an epitope image by a paratope image was achieved by using a convolutional neural network-based model, which was trained on a set of two-dimensional paratope-epitope images derived from experimental structures of antibody-antigen complexes. Our method achieves good performances in terms of cross-validation with a balanced accuracy of 0.8. Finally, we showcase examples of application of ImaPep, including extensive screening of large libraries to identify paratope candidates that bind to a selected epitope, and rescoring and refining antibody-antigen docking poses.

Animal-derived food allergen: A review on the available crystal structure and new insights into structural epitope

Immunoglobulin E (IgE)-mediated food allergy is a rapidly growing public health problem. The interaction between allergens and IgE is at the core of the allergic response. One of the best ways to understand this interaction is through structural characterization. This review focuses on animal-derived food allergens, overviews allergen structures determined by X-ray crystallography, presents an update on IgE conformational epitopes, and explores the structural features of these epitopes. The structural determinants of allergenicity and cross-reactivity are also discussed. Animal-derived food allergens are classified into limited protein families according to structural features, with the calcium-binding protein and actin-binding protein families dominating. Progress in epitope characterization has provided useful information on the structural properties of the IgE recognition region. The data reveals that epitopes are located in relatively protruding areas with negative surface electrostatic potential. Ligand binding and disulfide bonds are two intrinsic characteristics that influence protein structure and impact allergenicity. Shared structures, local motifs, and shared epitopes are factors that lead to cross-reactivity. The structural properties of epitope regions and structural determinants of allergenicity and cross-reactivity may provide directions for the prevention, diagnosis, and treatment of food allergies. Experimentally determined structure, especially that of antigen-antibody complexes, remains limited, and the identification of epitopes continues to be a bottleneck in the study of animal-derived food allergens. A combination of traditional immunological techniques and emerging bioinformatics technology will revolutionize how protein interactions are characterized.

High-resolution epitope mapping of commercial antibodies to ANCA antigens by yeast surface display

Epitope mapping provides critical insight into antibody-antigen interactions. Epitope mapping of autoantibodies from patients with autoimmune diseases can help elucidate disease immunogenesis and guide the development of antigen-specific therapies. Similarly, epitope mapping of commercial antibodies targeting known autoantigens enables the use of those antibodies to test specific hypotheses. Anti-Neutrophil Cytoplasmic Autoantibody (ANCA) vasculitis results from the formation of autoantibodies to multiple autoantigens, including myeloperoxidase (MPO), proteinase-3 (PR3), plasminogen (PLG), and peroxidasin (PXDN). To perform high-resolution epitope mapping of commercial antibodies to these autoantigens, we developed a novel yeast surface display library based on a series of >5000 overlapping peptides derived from their protein sequences. Using both FACS and magnetic bead isolation of reactive yeast, we screened 19 commercially available antibodies to the ANCA autoantigens. This approach to epitope mapping resulted in highly specific, fine epitope mapping, down to single amino acid resolution in many cases. Our study also identified cross-reactivity between some commercial antibodies to MPO and PXDN, which suggests that patients with apparent autoantibodies to both proteins may be the result of cross-reactivity. Together, our data validate yeast surface display using maximally overlapping peptides as an excellent approach to linear epitope mapping.

Linear epitopes of bony fish β-parvalbumins

Introduction

Fish β-parvalbumins are common targets of allergy-causing immunity. The nature of antibody responses to such allergens determines the biological outcome following exposure to fish. Specific epitopes on these allergens recognised by antibodies are incompletely characterised.

Methods

High-content peptide microarrays offer a solution to the identification of linear epitopes recognised by antibodies. We characterized IgG and IgG4 recognition of linear epitopes of fish β-parvalbumins defined in the WHO/IUIS allergen database as such responses hold the potential to counter an allergic reaction to these allergens. Peripheral blood samples, collected over three years, of 15 atopic but not fish-allergic subjects were investigated using a microarray platform that carried every possible 16-mer peptide of known isoforms and isoallergens of these and other allergens.

Results

Interindividual differences in epitope recognition patterns were observed. In contrast, reactivity patterns in a given individual were by comparison more stable during the 3 years-course of the study. Nevertheless, evidence of the induction of novel specificities over time was identified across multiple regions of the allergens. Particularly reactive epitopes were identified in the D helix of Cyp c 1 and in the C-terminus of Gad c 1 and Gad m 1.02. Residues important for the recognition of certain linear epitopes were identified. Patterns of differential recognition of isoallergens were observed in some subjects.

Conclusions

Altogether, comprehensive analysis of antibody recognition of linear epitopes of multiple allergens enables characterisation of the nature of the antibody responses targeting this important set of food allergens.

Integrating artificial intelligence-based epitope prediction in a SARS-CoV-2 antibody discovery pipeline: caution is warranted

BACKGROUND

SARS-CoV-2-neutralizing antibodies (nABs) showed great promise in the early phases of the COVID-19 pandemic. The emergence of resistant strains, however, quickly rendered the majority of clinically approved nABs ineffective. This underscored the imperative to develop nAB cocktails targeting non-overlapping epitopes.

METHODS

Undertaking a nAB discovery program, we employed a classical workflow, while integrating artificial intelligence (AI)-based prediction to select non-competing nABs very early in the pipeline. We identified and in vivo validated (in female Syrian hamsters) two highly potent nABs.

FINDINGS

Despite the promising results, in depth cryo-EM structural analysis demonstrated that the AI-based prediction employed with the intention to ensure non-overlapping epitopes was inaccurate. The two nABs in fact bound to the same receptor-binding epitope in a remarkably similar manner.

INTERPRETATION

Our findings indicate that, even in the Alphafold era, AI-based predictions of paratope-epitope interactions are rough and experimental validation of epitopes remains an essential cornerstone of a successful nAB lead selection.

FUNDING

Full list of funders is provided at the end of the manuscript.

Combined Immunoinformatics to Design and Evaluate a Multi-Epitope Vaccine Candidate against Streptococcus suis Infection

Streptococcus suis (S. suis) is a zoonotic pathogen with multiple serotypes, and thus, multivalent vaccines generating cross-protection against S. suis infections are urgently needed to improve animal welfare and reduce antibiotic abuse. In this study, we established a systematic and comprehensive epitope prediction pipeline based on immunoinformatics. Ten candidate epitopes were ultimately selected for building the multi-epitope vaccine (MVSS) against S. suis infections. The ten epitopes of MVSS were all derived from highly conserved, immunogenic, and virulence-associated surface proteins in S. suis. In silico analyses revealed that MVSS was structurally stable and affixed with immune receptors, indicating that it would likely trigger strong immunological reactions in the host. Furthermore, mice models demonstrated that MVSS elicited high titer antibodies and diminished damages in S. suis serotype 2 and Chz infection, significantly reduced sequelae, induced cytokine transcription, and decreased organ bacterial burdens after triple vaccination. Meanwhile, anti-rMVSS serum inhibited five important S. suis serotypes in vitro, exerted beneficial protective effects against S. suis infections and significantly reduced histopathological damage in mice. Given the above, it is possible to develop MVSS as a universal subunit vaccine against multiple serotypes of S. suis infections.

Nonclinical immunogenicity risk assessment for knobs-into-holes bispecific IgG1 antibodies

Bispecific antibodies, including bispecific IgG, are emerging as an important new class of antibody therapeutics. As a result, we, as well as others, have developed engineering strategies designed to facilitate the efficient production of bispecific IgG for clinical development. For example, we have extensively used knobs-into-holes (KIH) mutations to facilitate the heterodimerization of antibody heavy chains and more recently Fab mutations to promote cognate heavy/light chain pairing for efficient in vivo assembly of bispecific IgG in single host cells. A panel of related monospecific and bispecific IgG1 antibodies was constructed and assessed for immunogenicity risk by comparison with benchmark antibodies with known low (Avastin and Herceptin) or high (bococizumab and ATR-107) clinical incidence of anti-drug antibodies. Assay methods used include dendritic cell internalization, T cell proliferation, and T cell epitope identification by in silico prediction and MHC-associated peptide proteomics. Data from each method were considered independently and then together for an overall integrated immunogenicity risk assessment. In toto, these data suggest that the KIH mutations and in vitro assembly of half antibodies do not represent a major risk for immunogenicity of bispecific IgG1, nor do the Fab mutations used for efficient in vivo assembly of bispecifics in single host cells. Comparable or slightly higher immunogenicity risk assessment data were obtained for research-grade preparations of trastuzumab and bevacizumab versus Herceptin and Avastin, respectively. These data provide experimental support for the common practice of using research-grade preparations of IgG1 as surrogates for immunogenicity risk assessment of their corresponding pharmaceutical counterparts.

T and B cell epitope analysis for the immunogenicity evaluation and mitigation of antibody-based therapeutics

The surge in the clinical use of therapeutic antibodies has reshaped the landscape of pharmaceutical therapy for many diseases, including rare and challenging conditions. However, the administration of exogenous biologics could potentially trigger unwanted immune responses such as generation of anti-drug antibodies (ADAs). Real-world experiences have illuminated the clear correlation between the ADA occurrence and unsatisfactory therapeutic outcomes as well as immune-related adverse events. By retrospectively examining research involving immunogenicity analysis, we noticed the growing emphasis on elucidating the immunogenic epitope profiles of antibody-based therapeutics aiming for mechanistic understanding the immunogenicity generation and, ideally, mitigating the risks. As such, we have comprehensively summarized here the progress in both experimental and computational methodologies for the characterization of T and B cell epitopes of therapeutics. Furthermore, the successful practice of epitope-driven deimmunization of biotherapeutics is exceptionally highlighted in this article.

Limited value of current and new in silico predicted oocyst-specific proteins of Toxoplasma gondii for source-attributing serology

Toxoplasma gondii is a zoonotic parasite infecting all warm-blooded animals, including humans. The contribution of environmental contamination by T. gondii oocysts to infections is understudied. The aim of the current work was to explore T. gondii serology as a means of attributing the source of infection using a robust stepwise approach. We identified in silico thirty-two promising oocyst-specific antigens from T. gondii ´omics data, recombinantly expressed and purified them and validated whether serology based on these proteins could discriminate oocyst- from tissue cyst-driven experimental infections. For this, three well-characterized serum panels, sampled from 0 to 6 weeks post-infection, from pigs and sheep experimentally infected with T. gondii oocysts or tissue cysts, were used. Candidate proteins were initially screened by Western blot with sera from pigs or sheep, infected for different times, either with oocysts or tissue cysts, as well as non-infected animals. Only the recombinant proteins TgCCp5A and TgSR1 provoked seroconversion upon infection and appeared to discriminate between oocyst- and tissue cyst-driven infections with pig sera. They were subsequently used to develop an enzyme-linked immunosorbent assay test for pigs. Based on this assay and Western blot analyses, a lack of stage specificity and low antigenicity was observed with all pig sera. The same was true for proteins TgERP, TgSporoSAG, TgOWP1 and TgOWP8, previously described as source-attributing antigens, when analyzed using the whole panels of sera. We conclude that there is currently no antigen that allows the discrimination of T. gondii infections acquired from either oocysts or tissue cysts by serological tests. This work provides robust new knowledge that can inform further research and development toward source-attributing T. gondii serology.

An explainable language model for antibody specificity prediction using curated influenza hemagglutinin antibodies

Despite decades of antibody research, it remains challenging to predict the specificity of an antibody solely based on its sequence. Two major obstacles are the lack of appropriate models and inaccessibility of datasets for model training. In this study, we curated a dataset of >5,000 influenza hemagglutinin (HA) antibodies by mining research publications and patents, which revealed many distinct sequence features between antibodies to HA head and stem domains. We then leveraged this dataset to develop a lightweight memory B cell language model (mBLM) for sequence-based antibody specificity prediction. Model explainability analysis showed that mBLM captured key sequence motifs of HA stem antibodies. Additionally, by applying mBLM to HA antibodies with unknown epitopes, we discovered and experimentally validated many HA stem antibodies. Overall, this study not only advances our molecular understanding of antibody response to influenza virus, but also provides an invaluable resource for applying deep learning to antibody research.

Computational Methods in Immunology and Vaccinology: Design and Development of Antibodies and Immunogens

The design of new biomolecules able to harness immune mechanisms for the treatment of diseases is a prime challenge for computational and simulative approaches. For instance, in recent years, antibodies have emerged as an important class of therapeutics against a spectrum of pathologies. In cancer, immune-inspired approaches are witnessing a surge thanks to a better understanding of tumor-associated antigens and the mechanisms of their engagement or evasion from the human immune system. Here, we provide a summary of the main state-of-the-art computational approaches that are used to design antibodies and antigens, and in parallel, we review key methodologies for epitope identification for both B- and T-cell mediated responses. A special focus is devoted to the description of structure- and physics-based models, privileged over purely sequence-based approaches. We discuss the implications of novel methods in engineering biomolecules with tailored immunological properties for possible therapeutic uses. Finally, we highlight the extraordinary challenges and opportunities presented by the possible integration of structure- and physics-based methods with emerging Artificial Intelligence technologies for the prediction and design of novel antigens, epitopes, and antibodies.

Recent Progress in Antibody Epitope Prediction

Recent progress in epitope prediction has shown promising results in the development of vaccines and therapeutics against various diseases. However, the overall accuracy and success rate need to be improved greatly to gain practical application significance, especially conformational epitope prediction. In this review, we examined the general features of antibody-antigen recognition, highlighting the conformation selection mechanism in flexible antibody-antigen binding. We recently highlighted the success and warning signs of antibody epitope predictions, including linear and conformation epitope predictions. While deep learning-based models gradually outperform traditional feature-based machine learning, sequence and structure features still provide insight into antibody-antigen recognition problems.

Artificial intelligence in cancer immunotherapy: Applications in neoantigen recognition, antibody design and immunotherapy response prediction

Cancer immunotherapy is a method of controlling and eliminating tumors by reactivating the body's cancer-immunity cycle and restoring its antitumor immune response. The increased availability of data, combined with advancements in high-performance computing and innovative artificial intelligence (AI) technology, has resulted in a rise in the use of AI in oncology research. State-of-the-art AI models for functional classification and prediction in immunotherapy research are increasingly used to support laboratory-based experiments. This review offers a glimpse of the current AI applications in immunotherapy, including neoantigen recognition, antibody design, and prediction of immunotherapy response. Advancing in this direction will result in more robust predictive models for developing better targets, drugs, and treatments, and these advancements will eventually make their way into the clinical setting, pushing AI forward in the field of precision oncology.

Sph2(176-191) and Sph2(446-459): Identification of B-Cell Linear Epitopes in Sphingomyelinase 2 (Sph2), Naturally Recognized by Patients Infected by Pathogenic Leptospires

Sphingomyelin is a major constituent of eukaryotic cell membranes, and if degraded by bacteria sphingomyelinases may contribute to the pathogenesis of infection. Among Leptospira spp., there are five sphingomyelinases exclusively expressed by pathogenic leptospires, in which Sph2 is expressed during natural infections, cytotoxic, and implicated in the leptospirosis hemorrhagic complications. Considering this and the lack of information about associations between Sph2 and leptospirosis severity, we use a combination of immunoinformatics approaches to identify its B-cell epitopes, evaluate their reactivity against samples from leptospirosis patients, and investigate the role of antibodies anti-Sph2 in protection against severe leptospirosis. Two B-cell epitopes, Sph2_(176-191) and Sph2_(446-459), were predicted in Sph2 from L. interrogans serovar Lai, presenting different levels of identity when compared with other pathogenic leptospires. These epitopes were recognized by about 40% of studied patients with a prevalence of IgG antibodies against both Sph2_(176-191) and Sph2_(446-459). Remarkably, just individuals with low reactivity to Sph2_(176-191) presented clinical complications, while high responders had only mild symptoms. Therefore, we identified two B-cell linear epitopes, recognized by antibodies of patients with leptospirosis, that could be further explored in the development of multi-epitope vaccines against leptospirosis.