Computational design of SARS-CoV-2 spike glycoproteins to increase immunogenicity by T cell epitope engineering

CryoVirusDB: A Labeled Cryo-EM Image Dataset for AI-Driven Virus Particle Picking

With the advancements in instrumentation, image processing algorithms, and computational capabilities, single-particle electron cryo-microscopy (cryo-EM) has achieved nearly atomic resolution in determining the 3D structures of viruses. The virus structures play a crucial role in studying their biological function and advancing the development of antiviral vaccines and treatments. Despite the effectiveness of artificial intelligence (AI) in general image processing, its development for identifying and extracting virus particles from cryo-EM micrographs (images) has been hindered by the lack of manually labelled high-quality datasets. To fill the gap, we introduce CryoVirusDB, a labeled dataset containing the coordinates of expert-picked virus particles in cryo-EM micrographs. CryoVirusDB comprises 9,941 micrographs of 9 different viruses along with the coordinates of 339,398 labeled virus particles. It can be used to train and test AI and machine learning (e.g., deep learning) methods to accurately identify virus particles in cryo-EM micrographs for building atomic 3D structural models for viruses.

Vaxign-DL: A Deep Learning-based Method for Vaccine Design and its Evaluation

Reverse vaccinology (RV) provides a systematic approach to identifying potential vaccine candidates based on protein sequences. The integration of machine learning (ML) into this process has greatly enhanced our ability to predict viable vaccine candidates from these sequences. We have previously developed a Vaxign-ML program based on the eXtreme Gradient Boosting (XGBoost). In this study, we further extend our work to develop a Vaxign-DL program based on deep learning techniques. Deep neural networks assemble non-linear models and learn multilevel abstraction of data using hierarchically structured layers, offering a data-driven approach in computational design models. Vaxign-DL uses a three-layer fully connected neural network model. Using the same bacterial vaccine candidate training data as used in Vaxign-ML development, Vaxign-DL was able to achieve an Area Under the Receiver Operating Characteristic of 0.94, specificity of 0.99, sensitivity of 0.74, and accuracy of 0.96. Using the Leave-One-Pathogen-Out Validation (LOPOV) method, Vaxign-DL was able to predict vaccine candidates for 10 pathogens. Our benchmark study shows that Vaxign-DL achieved comparable results with Vaxign-ML in most cases, and our method outperforms Vaxi-DL in the accurate prediction of bacterial protective antigens.

From De Novo Design to Redesign: Harnessing Computational Protein Design for Understanding SARS-CoV-2 Molecular Mechanisms and Developing Therapeutics

The continuous emergence of novel SARS-CoV-2 variants and subvariants serves as compelling evidence that COVID-19 is an ongoing concern. The swift, well-coordinated response to the pandemic highlights how technological advancements can accelerate the detection, monitoring, and treatment of the disease. Robust surveillance systems have been established to understand the clinical characteristics of new variants, although the unpredictable nature of these variants presents significant challenges. Some variants have shown resistance to current treatments, but innovative technologies like computational protein design (CPD) offer promising solutions and versatile therapeutics against SARS-CoV-2. Advances in computing power, coupled with open-source platforms like AlphaFold and RFdiffusion (employing deep neural network and diffusion generative models), among many others, have accelerated the design of protein therapeutics with precise structures and intended functions. CPD has played a pivotal role in developing peptide inhibitors, mini proteins, protein mimics, decoy receptors, nanobodies, monoclonal antibodies, identifying drug-resistance mutations, and even redesigning native SARS-CoV-2 proteins. Pending regulatory approval, these designed therapies hold the potential for a lasting impact on human health and sustainability. As SARS-CoV-2 continues to evolve, use of such technologies enables the ongoing development of alternative strategies, thus equipping us for the "New Normal".

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.

Computational Protein Design for COVID-19 Research and Emerging Therapeutics

As the world struggles with the ongoing COVID-19 pandemic, unprecedented obstacles have continuously been traversed as new SARS-CoV-2 variants continually emerge. Infectious disease outbreaks are unavoidable, but the knowledge gained from the successes and failures will help create a robust health management system to deal with such pandemics. Previously, scientists required years to develop diagnostics, therapeutics, or vaccines; however, we have seen that, with the rapid deployment of high-throughput technologies and unprecedented scientific collaboration worldwide, breakthrough discoveries can be accelerated and insights broadened. Computational protein design (CPD) is a game-changing new technology that has provided alternative therapeutic strategies for pandemic management. In addition to the development of peptide-based inhibitors, miniprotein binders, decoys, biosensors, nanobodies, and monoclonal antibodies, CPD has also been used to redesign native SARS-CoV-2 proteins and human ACE2 receptors. We discuss how novel CPD strategies have been exploited to develop rationally designed and robust COVID-19 treatment strategies.

Investigation of the effects of N-Acetylglucosamine on the stability of the spike protein in SARS-CoV-2 by molecular dynamics simulations

A lot of effort has been made in developing vaccine and therapeutic agents against the SARS-CoV-2, concentrating on the Spike protein that binds angiotensin-converting enzyme 2 on human cells. Nowadays, some researches study the role of the N-linked glycans as potential targets for vaccines and new agents. Due to the flexibility and diversity of the N-linked glycans, in this work, we focus on the N-Acetylglucosamine moiety, which is the precursor of nearly all eukaryotic glycans. We performed molecular dynamics simulations to study the effects of the N-Acetylglucosamine on the stability of the spike glycoprotein in SARS-CoV-2. After a 100 ns of simulation on the spike proteins without and with the N-Acetylglucosamine molecules, we found that the presence of N-Acetylglucosamine increases the local stability in their vicinity; even though their effect on the full structure is negligible. Thus; it can be inferred that the N-Acetylglucosamine moieties can potentially affect the interaction of the S protein with the ACE2 receptor. We also found that the S1 domain is more flexible than the S2 domain. We propose which of the experimentally observed glycans found on the spike may be more functional than the others. Detailed understanding of glycans is key for the development of new therapeutic strategies.

COVID-19 vaccine design using reverse and structural vaccinology, ontology-based literature mining and machine learning

Rational vaccine design, especially vaccine antigen identification and optimization, is critical to successful and efficient vaccine development against various infectious diseases including coronavirus disease 2019 (COVID-19). In general, computational vaccine design includes three major stages: (i) identification and annotation of experimentally verified gold standard protective antigens through literature mining, (ii) rational vaccine design using reverse vaccinology (RV) and structural vaccinology (SV) and (iii) post-licensure vaccine success and adverse event surveillance and its usage for vaccine design. Protegen is a database of experimentally verified protective antigens, which can be used as gold standard data for rational vaccine design. RV predicts protective antigen targets primarily from genome sequence analysis. SV refines antigens through structural engineering. Recently, RV and SV approaches, with the support of various machine learning methods, have been applied to COVID-19 vaccine design. The analysis of post-licensure vaccine adverse event report data also provides valuable results in terms of vaccine safety and how vaccines should be used or paused. Ontology standardizes and incorporates heterogeneous data and knowledge in a human- and computer-interpretable manner, further supporting machine learning and vaccine design. Future directions on rational vaccine design are discussed.

Algorithms for protein design

Computational Protein Design has the potential to contribute to major advances in enzyme technology, vaccine design, receptor-ligand engineering, biomaterials, nanosensors, and synthetic biology. Although Protein Design is a challenging problem, proteins can be designed by experts in Protein Design, as well as by non-experts whose primary interests are in the applications of Protein Design. The increased accessibility of Protein Design technology is attributable to the accumulated knowledge and experience with Protein Design as well as to the availability of software and online resources. The objective of this review is to serve as a guide to the relevant literature with a focus on the novel methods and algorithms that have been developed or applied for Protein Design, and to assist in the selection of algorithms for Protein Design. Novel algorithms and models that have been introduced to utilize the enormous amount of experimental data and novel computational hardware have the potential for producing substantial increases in the accuracy, reliability and range of applications of designed proteins.

First computational design using lambda-superstrings and in vivo validation of SARS-CoV-2 vaccine

Coronavirus disease 2019 (COVID-19) is the greatest threat to global health at the present time, and considerable public and private effort is being devoted to fighting this recently emerged disease. Despite the undoubted advances in the development of vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of COVID-19, uncertainty remains about their future efficacy and the duration of the immunity induced. It is therefore prudent to continue designing and testing vaccines against this pathogen. In this article we computationally designed two candidate vaccines, one monopeptide and one multipeptide, using a technique involving optimizing lambda-superstrings, which was introduced and developed by our research group. We tested the monopeptide vaccine, thus establishing a proof of concept for the validity of the technique. We synthesized a peptide of 22 amino acids in length, corresponding to one of the candidate vaccines, and prepared a dendritic cell (DC) vaccine vector loaded with the 22 amino acids SARS-CoV-2 peptide (positions 50-71) contained in the NTD domain (DC-CoVPSA) of the Spike protein. Next, we tested the immunogenicity, the type of immune response elicited, and the cytokine profile induced by the vaccine, using a non-related bacterial peptide as negative control. Our results indicated that the CoVPSA peptide of the Spike protein elicits noticeable immunogenicity in vivo using a DC vaccine vector and remarkable cellular and humoral immune responses. This DC vaccine vector loaded with the NTD peptide of the Spike protein elicited a predominant Th1-Th17 cytokine profile, indicative of an effective anti-viral response. Finally, we performed a proof of concept experiment in humans that included the following groups: asymptomatic non-active COVID-19 patients, vaccinated volunteers, and control donors that tested negative for SARS-CoV-2. The positive control was the current receptor binding domain epitope of COVID-19 RNA-vaccines. We successfully developed a vaccine candidate technique involving optimizing lambda-superstrings and provided proof of concept in human subjects. We conclude that it is a valid method to decipher the best epitopes of the Spike protein of SARS-CoV-2 to prepare peptide-based vaccines for different vector platforms, including DC vaccines.

COVID-19 Vaccines: Current and Future Perspectives

Currently available vaccines against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) are highly effective but not able to keep the coronavirus disease 2019 (COVID-19) pandemic completely under control. Alternative R&D strategies are required to induce a long-lasting immunological response and to reduce adverse events as well as to favor rapid development and large-scale production. Several technological platforms have been used to develop COVID-19 vaccines, including inactivated viruses, recombinant proteins, DNA- and RNA-based vaccines, virus-vectored vaccines, and virus-like particles. In general, mRNA vaccines, protein-based vaccines, and vectored vaccines have shown a high level of protection against COVID-19. However, the mutation-prone nature of the spike (S) protein affects long-lasting vaccine protection and its effectiveness, and vaccinated people can become infected with new variants, also showing high virus levels. In addition, adverse effects may occur, some of them related to the interaction of the S protein with the angiotensin-converting enzyme 2 (ACE-2). Thus, there are some concerns that need to be addressed and challenges regarding logistic problems, such as strict storage at low temperatures for some vaccines. In this review, we discuss the limits of vaccines developed against COVID-19 and possible innovative approaches.

CD4+ T-cells from COVID-19 mRNA vaccine recipients recognize a conserved epitope present in diverse coronaviruses

Recent studies have shown that vaccinated individuals harbor T cells that can cross-recognize SARS-CoV-2 and endemic human common cold coronaviruses. However, it is still unknown whether CD4⁺ T cells from vaccinated individuals recognize peptides from bat coronaviruses that may have the potential of causing future pandemics. In this study, we identified a SARS-CoV-2 spike protein epitope (S_815-827) that is conserved in coronaviruses from different genera and subgenera, including SARS-CoV, MERS-CoV, multiple bat coronaviruses, and a feline coronavirus. Our results showed that S_815-827 was recognized by 42% of vaccinated participants in our study who received the Pfizer-BioNTech (BNT162b2) or Moderna (mRNA-1273) COVID-19 vaccines. Using T cell expansion and T cell receptor sequencing assays, we demonstrated that S_815-827-reactive CD4⁺ T cells from the majority of responders cross-recognized homologous peptides from at least 6 other diverse coronaviruses. Our results support the hypothesis that the current mRNA vaccines elicit T cell responses that can cross-recognize bat coronaviruses and thus might induce some protection against potential zoonotic outbreaks. Furthermore, our data provide important insights that inform the development of T cell–based pan-coronavirus vaccine strategies.

Severe acute respiratory syndrome coronavirus-2: An era of struggle and discovery leading to the emergency use authorization of treatment and prevention measures based on computational analysis

Severe acute respiratory syndrome-Coronavirus-2 (SARS-CoV-2), a novel betacoronavirus, has surprised the world with its disease spread and mortality rate. SARS-CoV-2 is a positive-sense, enveloped RNA virus that can infect various organs of the body, potentially leading to multiple organ dysfunction and eventual death. While various medications have received emergency use authorizations (EUAs) for the treatment of Coronavirus disease-2019 (COVID-19), as of April 30, 2021, only one drug has been Food and Drug Administration (FDA)-approved: remdesivir. Currently, three vaccines have received EUAs in the United States, but none are FDA-approved. This shortage of treatments and prevention measures is extremely problematic. Thus computational approaches would provide important data about drug resistance and variants. Such data will be useful for the development of drugs and vaccines. This chapter is a synopsis of SARS-CoV-2 clinical presentation, COVID-19 symptomology, treatment, prevention mechanisms, and SARS-CoV-2 variants using computational analysis.

SARS-CoV-2 Vaccines Based on the Spike Glycoprotein and Implications of New Viral Variants

Coronavirus 19 Disease (COVID-19) originating in the province of Wuhan, China in 2019, is caused by the severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), whose infection in humans causes mild or severe clinical manifestations that mainly affect the respiratory system. So far, the COVID-19 has caused more than 2 million deaths worldwide. SARS-CoV-2 contains the Spike (S) glycoprotein on its surface, which is the main target for current vaccine development because antibodies directed against this protein can neutralize the infection. Companies and academic institutions have developed vaccines based on the S glycoprotein, as well as its antigenic domains and epitopes, which have been proven effective in generating neutralizing antibodies. However, the emergence of new SARS-CoV-2 variants could affect the effectiveness of vaccines. Here, we review the different types of vaccines designed and developed against SARS-CoV-2, placing emphasis on whether they are based on the complete S glycoprotein, its antigenic domains such as the receptor-binding domain (RBD) or short epitopes within the S glycoprotein. We also review and discuss the possible effectiveness of these vaccines against emerging SARS-CoV-2 variants.

Lessons Learned from Cutting-Edge Immunoinformatics on Next-Generation COVID-19 Vaccine Research

Presently, immunoinformatics and bioinformatics approaches are contributing actively to COVID-19 vaccine research. The first immunoinformatics-based vaccine construct against SARS-CoV-2 was published in February 2020. Following this, immunoinformatics and bioinformatics approaches have created a new direction in COVID-19 vaccine research. Several researchers have designed the next-generation COVID-19 vaccines using these approaches. Presently, immunoinformatics has accelerated immunology research immensely in the area of COVID-19. Hence, we have tried to depict the current scenario of immunoinformatics and bioinformatics in COVID-19 vaccine research.

Vaxign2: the second generation of the first Web-based vaccine design program using reverse vaccinology and machine learning

Vaccination is one of the most significant inventions in medicine. Reverse vaccinology (RV) is a state-of-the-art technique to predict vaccine candidates from pathogen's genome(s). To promote vaccine development, we updated Vaxign2, the first web-based vaccine design program using reverse vaccinology with machine learning. Vaxign2 is a comprehensive web server for rational vaccine design, consisting of predictive and computational workflow components. The predictive part includes the original Vaxign filtering-based method and a new machine learning-based method, Vaxign-ML. The benchmarking results using a validation dataset showed that Vaxign-ML had superior prediction performance compared to other RV tools. Besides the prediction component, Vaxign2 implemented various post-prediction analyses to significantly enhance users' capability to refine the prediction results based on different vaccine design rationales and considerably reduce user time to analyze the Vaxign/Vaxign-ML prediction results. Users provide proteome sequences as input data, select candidates based on Vaxign outputs and Vaxign-ML scores, and perform post-prediction analysis. Vaxign2 also includes precomputed results from approximately 1 million proteins in 398 proteomes of 36 pathogens. As a demonstration, Vaxign2 was used to effectively analyse SARS-CoV-2, the coronavirus causing COVID-19. The comprehensive framework of Vaxign2 can support better and more rational vaccine design. Vaxign2 is publicly accessible at http://www.violinet.org/vaxign2.

Deep learning techniques have significantly impacted protein structure prediction and protein design

Protein structure prediction and design can be regarded as two inverse processes governed by the same folding principle. Although progress remained stagnant over the past two decades, the recent application of deep neural networks to spatial constraint prediction and end-to-end model training has significantly improved the accuracy of protein structure prediction, largely solving the problem at the fold level for single-domain proteins. The field of protein design has also witnessed dramatic improvement, where noticeable examples have shown that information stored in neural-network models can be used to advance functional protein design. Thus, incorporation of deep learning techniques into different steps of protein folding and design approaches represents an exciting future direction and should continue to have a transformative impact on both fields.

COVID-19 Vaccines: Current Understanding on Immunogenicity, Safety, and Further Considerations

The world has entered the second wave of the COVID-19 pandemic, and its intensity is significantly higher than that of the first wave of early 2020. Many countries or regions have been forced to start the second round of lockdowns. To respond rapidly to this global pandemic, dozens of COVID-19 vaccine candidates have been developed and many are undergoing clinical testing. Evaluating and defining effective vaccine candidates for human use is crucial for prioritizing vaccination programs against COVID-19. In this review, we have summarized and analyzed the efficacy, immunogenicity and safety data from clinical reports on different COVID-19 vaccines. We discuss the various guidelines laid out for the development of vaccines and the importance of biological standards for comparing the performance of vaccines. Lastly, we highlight the key remaining challenges, possible strategies for addressing them and the expected improvements in the next generation of COVID-19 vaccines.