RESUMEN

Arthropod-borne viruses, such as dengue virus (DENV), pose significant global health threats, with DENV alone infecting around 400 million people annually and causing outbreaks beyond endemic regions. This study aimed to enhance serological diagnosis and discover new drugs by identifying immunogenic protein regions of DENV. Utilizing a comprehensive approach, the study focused on peptides capable of distinguishing DENV from other flavivirus infections through serological analyses. Over 200 patients with confirmed arbovirus infection were profiled using high-density pan flavivirus peptide arrays comprising 6253 peptides and the computational method matrix of local coupling energy (MLCE). Twenty-four peptides from nonstructural and structural viral proteins were identified as specifically recognized by individuals with DENV infection. Six peptides were confirmed to distinguish DENV from Zika virus (ZIKV), West Nile virus (WNV), Yellow Fever virus (YFV), Usutu virus (USUV), and Chikungunya virus (CHIKV) infections, as well as healthy controls. Moreover, the combination of two immunogenic peptides emerged as a potential serum biomarker for DENV infection. These peptides, mapping to highly accessible regions on protein structures, show promise for diagnostic and prophylactic strategies against flavivirus infections. The described methodology holds broader applicability in the serodiagnosis of infectious diseases.

Asunto(s)

RESUMEN

Tropical theileriosis is a lymphoproliferative disease caused by Theileria annulata and is transmitted by Ixodid ticks of the genus Hyalomma. It causes significant losses in livestock, especially in exotic cattle. The existing methods for controlling it, chemotherapeutic agents and a vaccine based on an attenuated schizont stage parasite, have several limitations. A promising solution to control this disease is the use of molecular vaccines based on potential immunogenic proteins of T. annulata. For this purpose, we selected five antigenic sequences of T. annulata, i.e. SPAG-1, Tams, TaSP, spm2, and Ta9. These were subjected to epitope prediction for cytotoxic T lymphocytes, B-cells, and helper T lymphocytes. CTL and B-cell epitopes with a higher score whereas those of HTL with a lower score, were selected for the construct. A single protein was constructed using specific linkers and evaluated for high antigenicity and low allergenicity. The construct was acidic, hydrophobic, and thermostable in nature. Secondary and tertiary structures of this construct were drawn using the PSIPRED and RaptorX servers, respectively. A Ramachandran plot showed a high percentage of residues in this construct in favorable, allowed, and general regions. Molecular docking studies suggested that the complex was stable and our construct could potentially be a good candidate for immunization trials. Furthermore, we successfully cloned it into the pET-28a plasmid and transformed it into the BL21 strain. A restriction analysis was performed to confirm the transformation of our plasmid. After expression and purification, recombinant protein of 49 kDa was confirmed by western blotting. An ELISA detected increased specific antibody levels in the sera of the immunized animals compared with the control group, and flow cytometric analysis showed a stronger cell-mediated immune response. We believe our multi-epitope recombinant protein has the potential for the large-scale application for disease prevention globally in the bovine population. This study will act as a model for similar parasitic challenges.

Asunto(s)

RESUMEN

Brucellosis, a zoonotic disease caused by Brucella, presents a significant threat to both animal and human health. In animals, the disease can lead to infertility, miscarriage, and high fever, while in humans, symptoms may include recurrent fever, fatigue, sweating, hepatosplenomegaly, and joint and muscle pain following infection. Treatment often involves long-term antibiotic therapy, placing a substantial psychological and financial burden on patients. While vaccination is crucial for prevention, current animal vaccines have drawbacks such as residual virulence, and a safe and effective human vaccine is lacking. Hence, the development of a vaccine for brucellosis is imperative. In this study, we utilized bioinformatics methods to design a multi-epitope vaccine targeting Brucella. Targeting Heme transporter BhuA and polysaccharide export protein, we identified antigenic epitopes, including six cytotoxic T lymphocyte (CTL) dominant epitopes, six helper T lymphocyte (HTL) dominant epitopes, one conformation B cell dominant epitope, and three linear B cell dominant epitopes. By linking these epitopes with appropriate linkers and incorporating a Toll-like receptor (TLR) agonist (human beta-defensin-2) and an auxiliary peptide (Pan HLA-DR epitopes), we constructed the multi-epitope vaccine (MEV). The MEV demonstrated high antigenicity, non-toxicity, non-allergenicity, non-human homology, stability, and solubility. Molecular docking analysis and molecular dynamics simulations confirmed the interaction and stability of the MEV with receptors (MHCI, MHCII, TLR4). Codon optimization and in silico cloning validated the translation efficiency and successful expression of MEV in Escherichia coli. Immunological simulations further demonstrated the efficacy of MEV in inducing robust immune responses. In conclusion, our findings suggest that the engineered MEVs have the potential to stimulate both humoral and cellular immune responses, offering valuable insights for the future development of safe and efficient Brucella vaccines.

RESUMEN

Cancer cells and pathogens can evade T cell receptors (TCRs) via mutations in immunogenic epitopes. TCR cross-reactivity (i.e., recognition of multiple epitopes with sequence similarities) can counteract such escape but may cause severe side effects in cell-based immunotherapies through targeting self-antigens. To predict the effect of epitope point mutations on T cell functionality, we here present the random forest-based model Predicting T Cell Epitope-Specific Activation against Mutant Versions (P-TEAM). P-TEAM was trained and tested on three datasets with TCR responses to single-amino-acid mutations of the model epitope SIINFEKL, the tumor neo-epitope VPSVWRSSL, and the human cytomegalovirus antigen NLVPMVATV, totaling 9,690 unique TCR-epitope interactions. P-TEAM was able to accurately classify T cell reactivities and quantitatively predict T cell functionalities for unobserved single-point mutations and unseen TCRs. Overall, P-TEAM provides an effective computational tool to study T cell responses against mutated epitopes.

Asunto(s)

RESUMEN

BACKGROUND: High-risk human papillomavirus (HR-HPV) infection is an important factor for the development of cervical cancer. HPV18 is the second most common HR-HPV after HPV16. METHODS: In this study, MEGA11 software was used to analyze the variation and phylogenetic tree of HPV18 E6-E7 and L1 genes. The selective pressure to E6, E7 and L1 genes was estimated using pamlX. In addition, the B cell epitopes of L1 amino acid sequences and T cell epitopes of E6-E7 amino acid sequences in HPV18 were predicted by ABCpred server and IEDB website, respectively. RESULTS: A total of 9 single nucleotide variants were found in E6-E7 sequences, of which 2 were nonsynonymous variants and 7 were synonymous variants. Twenty single nucleotide variants were identified in L1 sequence, including 11 nonsynonymous variants and 9 synonymous variants. Phylogenetic analysis showed that E6-E7 and L1 sequences were all distributed in A lineage. In HPV18 E6, E7 and L1 sequences, no positively selected site was found. The nonconservative substitution R545C in L1 affected hypothetical B cell epitope. Two nonconservative substitutions, S82A in E6, and R53Q in E7, impacted multiple hypothetical T cell epitopes. CONCLUSION: The sequence variation data of HPV18 may lay a foundation for the virus diagnosis, further study of cervical cancer and vaccine design in central China.

Asunto(s)

RESUMEN

B-cell epitope prediction is key to developing peptide-based vaccines and immunodiagnostics along with antibodies for prophylactic, therapeutic and/or diagnostic use. This entails estimating paratope binding affinity for variable-length peptidic sequences subject to constraints on both paratope accessibility and antigen conformational flexibility, as described herein for the HAPTIC2/HEPTAD User Toolkit (HUT). HUT comprises the Heuristic Affinity Prediction Tool for Immune Complexes 2 (HAPTIC2), the HAPTIC2-like Epitope Prediction Tool for Antigen with Disulfide (HEPTAD) and the HAPTIC2/HEPTAD Input Preprocessor (HIP). HIP enables tagging of residues (e.g., in hydrophobic blobs, ordered regions and glycosylation motifs) for exclusion from downstream analyses by HAPTIC2 and HEPTAD. HAPTIC2 estimates paratope binding affinity for disulfide-free disordered peptidic antigens (by analogy between flexible-ligand docking and protein folding), from terms attributed to compaction (in view of sequence length, charge and temperature-dependent polyproline-II helical propensity), collapse (disfavored by residue bulkiness) and contact (with glycine and proline regarded as polar residues that hydrogen bond with paratopes). HEPTAD analyzes antigen sequences that each contain two cysteine residues for which the impact of disulfide pairing is estimated as a correction to the free-energy penalty of compaction. All of HUT is freely accessible online ( https://freeshell.de/~badong/hut.htm ).

Asunto(s)

RESUMEN

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.

RESUMEN

Tetanus disease, caused by C. tetani, starts with wounds or mucous layer contact. Prevented by vaccination, the lack of booster shots throughout life requires prophylactic treatment in case of accidents. The incidence of tetanus is high in underdeveloped countries, requiring the administration of antitetanus antibodies, usually derived from immunized horses or humans. Heterologous sera represent risks such as serum sickness. Human sera can carry unknown viruses. In the search for human monoclonal antibodies (mAbs) against TeNT (Tetanus Neurotoxin), we previously identified a panel of mAbs derived from B-cell sorting, selecting two nonrelated ones that binded to the C-terminal domain of TeNT (HCR/T), inhibiting its interaction with the cellular receptor ganglioside GT1b. Here, we present the results of cellular assays and molecular docking tools. TeNT internalization in neurons is prevented by more than 50% in neonatal rat spinal cord cells, determined by quantitative analysis of immunofluorescence punctate staining of Alexa Fluor 647 conjugated to TeNT. We also confirmed the mediator role of the Synaptic Vesicle Glycoprotein II (SV2) in TeNT endocytosis. The molecular docking assays to predict potential TeNT epitopes showed the binding of both antibodies to the HCR/T domain. A higher incidence was found between N1153 and W1297 when evaluating candidate residues for conformational epitope.

Asunto(s)

RESUMEN

The sensitivity of malignant tissues to T cell-based immunotherapies depends on the presence of targetable human leukocyte antigen (HLA) class I ligands. Peptide-intrinsic factors, such as HLA class I affinity and proteasomal processing, have been established as determinants of HLA ligand presentation. However, the role of gene and protein sequence features as determinants of epitope presentation has not been systematically evaluated. We perform HLA ligandome mass spectrometry to evaluate the contribution of 7,135 gene and protein sequence features to HLA sampling. This analysis reveals that a number of predicted modifiers of mRNA and protein abundance and turnover, including predicted mRNA methylation and protein ubiquitination sites, inform on the presence of HLA ligands. Importantly, integration of such "hard-coded" sequence features into a machine learning approach augments HLA ligand predictions to a comparable degree as experimental measures of gene expression. Our study highlights the value of gene and protein features for HLA ligand predictions.

Asunto(s)

RESUMEN

Identifying antigens within a pathogen is a critical task to develop effective vaccines and diagnostic methods, as well as understanding the evolution and adaptation to host immune responses. Historically, antigenicity was studied with experiments that evaluate the immune response against selected fragments of pathogens. Using this approach, the scientific community has gathered abundant information regarding which pathogenic fragments are immunogenic. The systematic collection of this data has enabled unraveling many of the fundamental rules underlying the properties defining epitopes and immunogenicity, and has resulted in the creation of a large panel of immunologically relevant predictive (in silico) tools. The development and application of such tools have proven to accelerate the identification of novel epitopes within biomedical applications reducing experimental costs. This chapter introduces some basic concepts about MHC presentation, T cell and B cell epitopes, the experimental efforts to determine those, and focuses on state-of-the-art methods for epitope prediction, highlighting their strengths and limitations, and catering instructions for their rational use.

Asunto(s)

RESUMEN

BACKGROUND: Alpha-papillomavirus 9 (α-9) is a member of the human papillomavirus (HPV) α genus, causing 75% invasive cervical cancers worldwide. The purpose of this study was to provide data for effective treatment of HPV-induced cervical lesions in Taizhou by analysing the genetic variation and antigenic epitopes of α-9 HPV E6 and E7. METHODS: Cervical exfoliated cells were collected for HPV genotyping. Positive samples of the α-9 HPV single type were selected for E6 and E7 gene sequencing. The obtained nucleotide sequences were translated into amino acid sequences (protein primary structure) using MEGA X, and positive selection sites of the amino acid sequences were evaluated using PAML. The secondary and tertiary structures of the E6 and E7 proteins were predicted using PSIPred, SWISS-MODEL, and PyMol. Potential T/B-cell epitopes were predicted by Industrial Engineering Database (IEDB). RESULTS: From 2012 to 2023, α-9 HPV accounted for 75.0% (7815/10423) of high-risk HPV-positive samples in Taizhou, both alone and in combination with other types. Among these, single-type-positive samples of α-9 HPV were selected, and the entire E6 and E7 genes were sequenced, including 298 HPV16, 149 HPV31, 185 HPV33, 123 HPV35, 325 HPV52, and 199 HPV58 samples. Compared with reference sequences, 34, 12, 10, 2, 17, and 17 nonsynonymous nucleotide mutations were detected in HPV16, 31, 33, 35, 52, and 58, respectively. Among all nonsynonymous nucleotide mutations, 19 positive selection sites were selected, which may have evolutionary significance in rendering α-9 HPV adaptive to its environment. Immunoinformatics predicted 57 potential linear and 59 conformational B-cell epitopes, many of which are also predicted as CTL epitopes. CONCLUSION: The present study provides almost comprehensive data on the genetic variations, phylogenetics, positive selection sites, and antigenic epitopes of α-9 HPV E6 and E7 in Taizhou, China, which will be helpful for local HPV therapeutic vaccine development.

Asunto(s)

RESUMEN

B-cell epitope identification plays a vital role in the development of vaccines, therapies, and diagnostic tools. Currently, molecular docking tools in B-cell epitope prediction are heavily influenced by empirical parameters and require significant computational resources, rendering a great challenge to meet large-scale prediction demands. When predicting epitopes from antigen-antibody complex, current artificial intelligence algorithms cannot accurately implement the prediction due to insufficient protein feature representations, indicating novel algorithm is desperately needed for efficient protein information extraction. In this paper, we introduce a multimodal model called WUREN (Whole-modal Union Representation for Epitope predictioN), which effectively combines sequence, graph, and structural features. It achieved AUC-PR scores of 0.213 and 0.193 on the solved structures and AlphaFold-generated structures, respectively, for the independent test proteins selected from DiscoTope3 benchmark. Our findings indicate that WUREN is an efficient feature extraction model for protein complexes, with the generalizable application potential in the development of protein-based drugs. Moreover, the streamlined framework of WUREN could be readily extended to model similar biomolecules, such as nucleic acids, carbohydrates, and lipids.

RESUMEN

The development of vaccines against a wide range of infectious diseases and pathogens often relies on multi-epitope strategies that can effectively stimulate both humoral and cellular immunity. Immunoinformatics tools play a pivotal role in designing such vaccines, enhancing immune response potential, and minimizing the risk of failure. This review presents a comprehensive overview of practical tools for epitope prediction and the associated immune responses. These immunoinformatics tools facilitate the selection of epitopes based on parameters such as antigenicity, absence of toxic and allergenic sequences, secondary and tertiary structures, sequence conservation, and population coverage. The chosen epitopes can be tailored for B-cells or T-cells, both of which require further assessments covered in this study. We offer a range of suitable linkers that effectively separate cytotoxic T lymphocyte and helper T lymphocyte epitopes while preserving their functionality. Additionally, we identify various adjuvants for specific purposes. We delve into the evaluation of MHC-epitope interactions, MHC clusters, and the simulation of final constructs through molecular docking techniques. We provide diverse linkers and adjuvants optimized for epitope functions to bolster immune responses through epitope attachment. By leveraging these comprehensive tools, the development of multi-epitope vaccines holds the promise of robust immunity and a significant reduction in experimental costs.

Asunto(s)

RESUMEN

Defining the binding epitopes of antibodies is essential for understanding how they bind to their antigens and perform their molecular functions. However, while determining linear epitopes of monoclonal antibodies can be accomplished utilizing well-established empirical procedures, these approaches are generally labor- and time-intensive and costly. To take advantage of the recent advances in protein structure prediction algorithms available to the scientific community, we developed a calculation pipeline based on the localColabFold implementation of AlphaFold2 that can predict linear antibody epitopes by predicting the structure of the complex between antibody heavy and light chains and target peptide sequences derived from antigens. We found that this AlphaFold2 pipeline, which we call PAbFold, was able to accurately flag known epitope sequences for several well-known antibody targets (HA / Myc) when the target sequence was broken into small overlapping linear peptides and antibody complementarity determining regions (CDRs) were grafted onto several different antibody framework regions in the single-chain antibody fragment (scFv) format. To determine if this pipeline was able to identify the epitope of a novel antibody with no structural information publicly available, we determined the epitope of a novel anti-SARS-CoV-2 nucleocapsid targeted antibody using our method and then experimentally validated our computational results using peptide competition ELISA assays. These results indicate that the AlphaFold2-based PAbFold pipeline we developed is capable of accurately identifying linear antibody epitopes in a short time using just antibody and target protein sequences. This emergent capability of the method is sensitive to methodological details such as peptide length, AlphaFold2 neural network versions, and multiple-sequence alignment database. PAbFold is available at https://github.com/jbderoo/PAbFold.

RESUMEN

Mycoplasma genitalium is a pathogenic microorganism linked to a variety of severe health conditions including ovarian cancer, prostate cancer, HIV transmission, and sexually transmitted diseases. A more effective approach to address the challenges posed by this pathogen, given its high antibiotic resistance rates, could be the development of a peptide vaccine. In this study, we used experimentally validated 13 membrane proteins and their immunogenicity to identify suitable vaccine candidates. Thus, based on immunogenic properties and high conservation among other Mycoplasma genitalium sub-strains, the P110 surface protein is considered for further investigation. Later on, we identified T-cell epitopes and B-cell epitopes from the P110 protein to construct a multiepitope-based vaccine. As a result, the 'NIAPISFSFTPFTAA' T-cell epitope and 'KVKYESSGSNNISFDS' B-cell epitope have shown 99.53% and 87.50% population coverage along with 100% conservancy among the subspecies, and both epitopes were found to be non-allergenic. Furthermore, focusing on molecular docking analysis showed the lowest binding energy for MHC-I (-137.5 kcal/mol) and MHC-II (-183.3 kcal/mol), leading to a satisfactory binding strength between the T-cell epitopes and the MHC molecules. However, the constructed multiepitope vaccine (MEV) consisting of 54 amino acids demonstrates favorable characteristics for a vaccine candidate, including a theoretical pI of 4.25 with a scaled solubility of 0.812 and high antigenicity probabilities. Additionally, structural analyses reveal that the MEV displays substantial alpha helices and extended strands, vital for its immunogenicity. Molecular docking with the human Toll-like receptors TLR1/2 heterodimer shows strong binding affinity, reinforcing its potential to elicit an immune response. Our immune simulation analysis demonstrates immune memory development and robust immunity, while codon adaptation suggests optimal expression in E. coli using the pET-28a(+) vector. These findings collectively highlight the MEV's potential as a valuable vaccine candidate against M. genitalium.

RESUMEN

Human papillomavirus type 16 (HPV16) infection is responsible for more than 50% of global cervical cancer cases. The development of a vaccine based on cytotoxic T-lymphocyte (CTL) epitopes is a promising strategy for eliminating pre-existing HPV infections and treating patients with cervical cancer. In this study, an immunoinformatics approach was used to predict HLA-I-restricted CTL epitopes in HPV16 E5, E6, and E7 proteins, and a set of conserved CTL epitopes co-restricted by human/murine MHCs was screened and characterized, with the set containing three E5, four E6, and four E7 epitopes. Subsequently, the immunogenicity of the epitope combination was assessed in mice, and the anti-tumor effects of the multi-epitope peptide vaccine E5E6E7pep11 and the recombinant protein vaccine CTB-Epi11E567 were evaluated in the TC-1 mouse tumor model. The results demonstrated that mixed epitope peptides could induce antigen-specific IFN-γ secretion in mice. Prophylactic immunization with E5E6E7pep11 and CTB-Epi11E567 was found to provide 100% protection against tumor growth in mice. Moreover, both types of the multi-epitope vaccine significantly inhibited tumor growth and prolonged mouse survival. In conclusion, in this study, a multi-epitope vaccine targeting HPV16 E5, E6, and E7 proteins was successfully designed and evaluated, demonstrating potential immunogenicity and anti-tumor effects and providing a promising strategy for immunotherapy against HPV-associated tumors.

RESUMEN

Vaccines have significantly reduced the impact of numerous deadly viral infections. However, there is an increasing need to expedite vaccine development in light of the recurrent pandemics and epidemics. Also, identifying vaccines against certain viruses is challenging due to various factors, notably the inability to culture certain viruses in cell cultures and the wide-ranging diversity of MHC profiles in humans. Fortunately, reverse vaccinology (RV) efficiently overcomes these limitations and has simplified the identification of epitopes from antigenic proteins across the entire proteome, streamlining the vaccine development process. Furthermore, it enables the creation of multiepitope vaccines that can effectively account for the variations in MHC profiles within the human population. The RV approach offers numerous advantages in developing precise and effective vaccines against viral pathogens, including extensive proteome coverage, accurate epitope identification, cross-protection capabilities, and MHC compatibility. With the introduction of RV, there is a growing emphasis among researchers on creating multiepitope-based vaccines aiming to stimulate the host's immune responses against multiple serotypes, as opposed to single-component monovalent alternatives. Regardless of how promising the RV-based vaccine candidates may appear, they must undergo experimental validation to probe their protection efficacy for real-world applications. The time, effort, and resources allocated to the laborious epitope identification process can now be redirected toward validating vaccine candidates identified through the RV approach. However, to overcome failures in the RV-based approach, efforts must be made to incorporate immunological principles and consider targeting the epitope regions involved in disease pathogenesis, immune responses, and neutralizing antibody maturation. Integrating multi-omics and incorporating artificial intelligence and machine learning-based tools and techniques in RV would increase the chances of developing an effective vaccine. This review thoroughly explains the RV approach, ideal RV-based vaccine construct components, RV-based vaccines designed to combat viral pathogens, its challenges, and future perspectives.

Asunto(s)

RESUMEN

Computational predictors of immunogenic peptides, or epitopes, are traditionally built based on data from a broad range of pathogens without consideration for taxonomic information. While this approach may be reasonable if one aims to develop one-size-fits-all models, it may be counterproductive if the proteins for which the model is expected to generalize are known to come from a specific subset of phylogenetically related pathogens. There is mounting evidence that, for these cases, taxon-specific models can outperform generalist ones, even when trained with substantially smaller amounts of data. In this comment, we provide some perspective on the current state of taxon-specific modelling for the prediction of linear B-cell epitopes, and the challenges faced when building and deploying these predictors.

Asunto(s)

RESUMEN

B cell epitope prediction methods are separated into linear sequence-based predictors and conformational epitope predictions that typically use the measured or predicted protein structure. Most linear predictions rely on the translation of the sequence to biologically based representations and the applications of machine learning on these representations. We here present CALIBER 'Conformational And LInear B cell Epitopes pRediction', and show that a bidirectional long short-term memory with random projection produces a more accurate prediction (test set AUC=0.789) than all current linear methods. The same predictor when combined with an Evolutionary Scale Modeling-2 projection also improves on the state of the art in conformational epitopes (AUC = 0.776). The inclusion of the graph of the 3D distances between residues did not increase the prediction accuracy. However, the long-range sequence information was essential for high accuracy. While the same model structure was applicable for linear and conformational epitopes, separate training was required for each. Combining the two slightly increased the linear accuracy (AUC 0.775 versus 0.768) and reduced the conformational accuracy (AUC = 0.769).

Asunto(s)

RESUMEN

Accurate computational identification of B-cell epitopes is crucial for the development of vaccines, therapies, and diagnostic tools. However, current structure-based prediction methods face limitations due to the dependency on experimentally solved structures. Here, we introduce DiscoTope-3.0, a markedly improved B-cell epitope prediction tool that innovatively employs inverse folding structure representations and a positive-unlabelled learning strategy, and is adapted for both solved and predicted structures. Our tool demonstrates a considerable improvement in performance over existing methods, accurately predicting linear and conformational epitopes across multiple independent datasets. Most notably, DiscoTope-3.0 maintains high predictive performance across solved, relaxed and predicted structures, alleviating the need for experimental structures and extending the general applicability of accurate B-cell epitope prediction by 3 orders of magnitude. DiscoTope-3.0 is made widely accessible on two web servers, processing over 100 structures per submission, and as a downloadable package. In addition, the servers interface with RCSB and AlphaFoldDB, facilitating large-scale prediction across over 200 million cataloged proteins. DiscoTope-3.0 is available at: https://services.healthtech.dtu.dk/service.php?DiscoTope-3.0.

Asunto(s)