Structure and inhibition of SARS-CoV-2 spike refolding in membranes.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein binds the receptor angiotensin converting enzyme 2 (ACE2) and drives virus-host membrane fusion through refolding of its S2 domain. Whereas the S1 domain contains high sequence variability, the S2 domain is conserved and is a promising pan-betacoronavirus vaccine target. We applied cryo-electron tomography to capture intermediates of S2 refolding and understand inhibition by antibodies to the S2 stem-helix. Subtomogram averaging revealed ACE2 dimers cross-linking spikes before transitioning into S2 intermediates, which were captured at various stages of refolding. Pan-betacoronavirus neutralizing antibodies targeting the S2 stem-helix bound to and inhibited refolding of spike prehairpin intermediates. Combined with molecular dynamics simulations, these structures elucidate the process of SARS-CoV-2 entry and reveal how pan-betacoronavirus S2-targeting antibodies neutralize infectivity by arresting prehairpin intermediates.

The alarmin IL-33 exacerbates pulmonary inflammation and immune dysfunction in SARS-CoV-2 infection.

Dysregulated host immune responses contribute to disease severity and worsened prognosis in COVID-19 infection and the underlying mechanisms are not fully understood. In this study, we observed that IL-33, a damage-associated molecular pattern molecule, is significantly increased in COVID-19 patients and in SARS-CoV-2-infected mice. Using IL-33-/- mice, we demonstrated that IL-33 deficiency resulted in significant decreases in bodyweight loss, tissue viral burdens, and lung pathology. These improved outcomes in IL-33-/- mice also correlated with a reduction in innate immune cell infiltrates, i.e., neutrophils, macrophages, natural killer cells, and activated T cells in inflamed lungs. Lung RNA-seq results revealed that IL-33 signaling enhances activation of inflammatory pathways, including interferon signaling, pathogen phagocytosis, macrophage activation, and cytokine/chemokine signals. Overall, these findings demonstrate that the alarmin IL-33 plays a pathogenic role in SARS-CoV-2 infection and provides new insights that will inform the development of effective therapeutic strategies for COVID-19.

TRIM7 ubiquitinates SARS-CoV-2 membrane protein to limit apoptosis and viral replication.

SARS-CoV-2 is a highly transmissible virus that causes COVID-19 disease. Mechanisms of viral pathogenesis include excessive inflammation and viral-induced cell death, resulting in tissue damage. We identified the host E3-ubiquitin ligase TRIM7 as an inhibitor of apoptosis and SARS-CoV-2 replication via ubiquitination of the viral membrane (M) protein. Trim7 -/- mice exhibited increased pathology and virus titers associated with epithelial apoptosis and dysregulated immune responses. Mechanistically, TRIM7 ubiquitinates M on K14, which protects cells from cell death. Longitudinal SARS-CoV-2 sequence analysis from infected patients revealed that mutations on M-K14 appeared in circulating variants during the pandemic. The relevance of these mutations was tested in a mouse model. A recombinant M-K14/K15R virus showed reduced viral replication, consistent with the role of K15 in virus assembly, and increased levels of apoptosis associated with the loss of ubiquitination on K14. TRIM7 antiviral activity requires caspase-6 inhibition, linking apoptosis with viral replication and pathology.

Lessons learnt from broad-spectrum coronavirus antiviral drug discovery.

INTRODUCTION: Highly pathogenic coronaviruses (CoVs), such as severe acute respiratory syndrome CoV (SARS-CoV), Middle East respiratory syndrome CoV (MERS-CoV), and the most recent SARS-CoV-2 responsible for the COVID-19 pandemic, pose significant threats to human populations over the past two decades. These CoVs have caused a broad spectrum of clinical manifestations ranging from asymptomatic to severe distress syndromes (ARDS), resulting in high morbidity and mortality. AREAS COVERED: The accelerated advancements in antiviral drug discovery, spurred by the COVID-19 pandemic, have shed new light on the imperative to develop treatments effective against a broad spectrum of CoVs. This perspective discusses strategies and lessons learnt in targeting viral non-structural proteins, structural proteins, drug repurposing, and combinational approaches for the development of antivirals against CoVs. EXPERT OPINION: Drawing lessons from the pandemic, it becomes evident that the absence of efficient broad-spectrum antiviral drugs increases the vulnerability of public health systems to the potential onslaught by highly pathogenic CoVs. The rapid and sustained spread of novel CoVs can have devastating consequences without effective and specifically targeted treatments. Prioritizing the effective development of broad-spectrum antivirals is imperative for bolstering the resilience of public health systems and mitigating the potential impact of future highly pathogenic CoVs.

Discovery of First-in-Class PROTAC Degraders of SARS-CoV-2 Main Protease.

We have witnessed three coronavirus (CoV) outbreaks in the past two decades, including the COVID-19 pandemic caused by SARS-CoV-2. Main protease (MPro), a highly conserved protease among various CoVs, is essential for viral replication and pathogenesis, making it a prime target for antiviral drug development. Here, we leverage proteolysis targeting chimera (PROTAC) technology to develop a new class of small-molecule antivirals that induce the degradation of SARS-CoV-2 MPro. Among them, MPD2 was demonstrated to effectively reduce MPro protein levels in 293T cells, relying on a time-dependent, CRBN-mediated, and proteasome-driven mechanism. Furthermore, MPD2 exhibited remarkable efficacy in diminishing MPro protein levels in SARS-CoV-2-infected A549-ACE2 cells. MPD2 also displayed potent antiviral activity against various SARS-CoV-2 strains and exhibited enhanced potency against nirmatrelvir-resistant viruses. Overall, this proof-of-concept study highlights the potential of targeted protein degradation of MPro as an innovative approach for developing antivirals that could fight against drug-resistant viral variants.

Resistance mechanisms of SARS-CoV-2 3CLpro to the non-covalent inhibitor WU-04.

Drug resistance poses a significant challenge in the development of effective therapies against SARS-CoV-2. Here, we identified two double mutations, M49K/M165V and M49K/S301P, in the 3C-like protease (3CLpro) that confer resistance to a novel non-covalent inhibitor, WU-04, which is currently in phase III clinical trials (NCT06197217). Crystallographic analysis indicates that the M49K mutation destabilizes the WU-04-binding pocket, impacting the binding of WU-04 more significantly than the binding of 3CLpro substrates. The M165V mutation directly interferes with WU-04 binding. The S301P mutation, which is far from the WU-04-binding pocket, indirectly affects WU-04 binding by restricting the rotation of 3CLpro's C-terminal tail and impeding 3CLpro dimerization. We further explored 3CLpro mutations that confer resistance to two clinically used inhibitors: ensitrelvir and nirmatrelvir, and revealed a trade-off between the catalytic activity, thermostability, and drug resistance of 3CLpro. We found that mutations at the same residue (M49) can have distinct effects on the 3CLpro inhibitors, highlighting the importance of developing multiple antiviral agents with different skeletons for fighting SARS-CoV-2. These findings enhance our understanding of SARS-CoV-2 resistance mechanisms and inform the development of effective therapeutics.

Targeting G9a translational mechanism of SARS-CoV-2 pathogenesis for multifaceted therapeutics of COVID-19 and its sequalae.

By largely unknown mechanism(s), SARS-CoV-2 hijacks the host translation apparatus to promote COVID-19 pathogenesis. We report that the histone methyltransferase G9a noncanonically regulates viral hijacking of the translation machinery to bring about COVID-19 symptoms of hyperinflammation, lymphopenia, and blood coagulation. Chemoproteomic analysis of COVID-19 patient peripheral mononuclear blood cells (PBMC) identified enhanced interactions between SARS-CoV-2-upregulated G9a and distinct translation regulators, particularly the N 6 -methyladenosine (m 6 A) RNA methylase METTL3. These interactions with translation regulators implicated G9a in translational regulation of COVID-19. Inhibition of G9a activity suppressed SARS-CoV-2 replication in human alveolar epithelial cells. Accordingly, multi-omics analysis of the same alveolar cells identified SARS-CoV-2-induced changes at the transcriptional, m 6 A-epitranscriptional, translational, and post-translational (phosphorylation or secretion) levels that were reversed by inhibitor treatment. As suggested by the aforesaid chemoproteomic analysis, these multi-omics-correlated changes revealed a G9a-regulated translational mechanism of COVID-19 pathogenesis in which G9a directs translation of viral and host proteins associated with SARS-CoV-2 replication and with dysregulation of host response. Comparison of proteomic analyses of G9a inhibitor-treated, SARS-CoV-2 infected cells, or ex vivo culture of patient PBMCs, with COVID-19 patient data revealed that G9a inhibition reversed the patient proteomic landscape that correlated with COVID-19 pathology/symptoms. These data also indicated that the G9a-regulated, inhibitor-reversed, translational mechanism outperformed G9a-transcriptional suppression to ultimately determine COVID-19 pathogenesis and to define the inhibitor action, from which biomarkers of serve symptom vulnerability were mechanistically derived. This cell line-to-patient conservation of G9a-translated, COVID-19 proteome suggests that G9a inhibitors can be used to treat patients with COVID-19, particularly patients with long-lasting COVID-19 sequelae.

Safety and Immunogenicity of the Monovalent Omicron XBB.1.5-Adapted BNT162b2 COVID-19 Vaccine in Individuals ≥12 Years Old: A Phase 2/3 Trial.

Vaccination remains an important mitigation tool against COVID-19. We report 1-month safety and preliminary immunogenicity data from a substudy of an ongoing, open-label, phase 2/3 study of monovalent Omicron XBB.1.5-adapted BNT162b2 (single 30-µg dose). Healthy participants ≥12 years old (N = 412 (12-17 years, N = 30; 18-55 years, N = 174; >55 years, N = 208)) who previously received ≥3 doses of a US-authorized mRNA vaccine, the most recent being an Omicron BA.4/BA.5-adapted bivalent vaccine ≥150 days before study vaccination, were vaccinated. Serum 50% neutralizing titers against Omicron XBB.1.5, EG.5.1, and BA.2.86 were measured 7 days and 1 month after vaccination in a subset of ≥18-year-olds (N = 40) who were positive for SARS-CoV-2 at baseline. Seven-day immunogenicity was also evaluated in a matched group who received bivalent BA.4/BA.5-adapted BNT162b2 in a previous study (ClinicalTrials.gov Identifier: NCT05472038). There were no new safety signals; local reactions and systemic events were mostly mild to moderate in severity, adverse events were infrequent, and none led to study withdrawal. The XBB.1.5-adapted BNT162b2 induced numerically higher titers against Omicron XBB.1.5, EG.5.1, and BA.2.86 than BA.4/BA.5-adapted BNT162b2 at 7 days and robust neutralizing responses to all three sublineages at 1 month. These data support a favorable benefit-risk profile of XBB.1.5-adapted BNT162b2 30 µg. ClinicalTrials.gov Identifier: NCT05997290.

A Survey of Deep Learning for Detecting miRNA- Disease Associations: Databases, Computational Methods, Challenges, and Future Directions.

MicroRNAs (miRNAs) are an important class of non-coding RNAs that play an essential role in the occurrence and development of various diseases. Identifying the potential miRNA-disease associations (MDAs) can be beneficial in understanding disease pathogenesis. Traditional laboratory experiments are expensive and time-consuming. Computational models have enabled systematic large-scale prediction of potential MDAs, greatly improving the research efficiency. With recent advances in deep learning, it has become an attractive and powerful technique for uncovering novel MDAs. Consequently, numerous MDA prediction methods based on deep learning have emerged. In this review, we first summarize publicly available databases related to miRNAs and diseases for MDA prediction. Next, we outline commonly used miRNA and disease similarity calculation and integration methods. Then, we comprehensively review the 48 existing deep learning-based MDA computation methods, categorizing them into classical deep learning and graph neural network-based techniques. Subsequently, we investigate the evaluation methods and metrics that are frequently used to assess MDA prediction performance. Finally, we discuss the performance trends of different computational methods, point out some problems in current research, and propose 9 potential future research directions. Data resources and recent advances in MDA prediction methods are summarized in the GitHub repository https://github.com/sheng-n/DL-miRNA-disease-association-methods.

Integrated multi-omics analyses identify anti-viral host factors and pathways controlling SARS-CoV-2 infection.

Host anti-viral factors are essential for controlling SARS-CoV-2 infection but remain largely unknown due to the biases of previous large-scale studies toward pro-viral host factors. To fill in this knowledge gap, we perform a genome-wide CRISPR dropout screen and integrate analyses of the multi-omics data of the CRISPR screen, genome-wide association studies, single-cell RNA-Seq, and host-virus proteins or protein/RNA interactome. This study uncovers many host factors that are currently underappreciated, including the components of V-ATPases, ESCRT, and N-glycosylation pathways that modulate viral entry and/or replication. The cohesin complex is also identified as an anti-viral pathway, suggesting an important role of three-dimensional chromatin organization in mediating host-viral interaction. Furthermore, we discover another anti-viral regulator KLF5, a transcriptional factor involved in sphingolipid metabolism, which is up-regulated, and harbors genetic variations linked to COVID-19 patients with severe symptoms. Anti-viral effects of three identified candidates (DAZAP2/VTA1/KLF5) are confirmed individually. Molecular characterization of DAZAP2/VTA1/KLF5-knockout cells highlights the involvement of genes related to the coagulation system in determining the severity of COVID-19. Together, our results provide further resources for understanding the host anti-viral network during SARS-CoV-2 infection and may help develop new countermeasure strategies.

Dengue and Zika RNA-RNA interactomes reveal pro- and anti-viral RNA in human cells.

BACKGROUND: Identifying host factors is key to understanding RNA virus pathogenicity. Besides proteins, RNAs can interact with virus genomes to impact replication. RESULTS: Here, we use proximity ligation sequencing to identify virus-host RNA interactions for four strains of Zika virus (ZIKV) and one strain of dengue virus (DENV-1) in human cells. We find hundreds of coding and non-coding RNAs that bind to DENV and ZIKV viruses. Host RNAs tend to bind to single-stranded regions along the virus genomes according to hybridization energetics. Compared to SARS-CoV-2 interactors, ZIKV-interacting host RNAs tend to be downregulated upon virus infection. Knockdown of several short non-coding RNAs, including miR19a-3p, and 7SK RNA results in a decrease in viral replication, suggesting that they act as virus-permissive factors. In addition, the 3'UTR of DYNLT1 mRNA acts as a virus-restrictive factor by binding to the conserved dumbbell region on DENV and ZIKV 3'UTR to decrease virus replication. We also identify a conserved set of host RNAs that interacts with DENV, ZIKV, and SARS-CoV-2, suggesting that these RNAs are broadly important for RNA virus infection. CONCLUSIONS: This study demonstrates that host RNAs can impact virus replication in permissive and restrictive ways, expanding our understanding of host factors and RNA-based gene regulation during viral pathogenesis.

SARS-COV-2 Omicron variants conformationally escape a rare quaternary antibody binding mode.

The ongoing evolution of SARS-CoV-2 into more easily transmissible and infectious variants has provided unprecedented insight into mutations enabling immune escape. Understanding how these mutations affect the dynamics of antibody-antigen interactions is crucial to the development of broadly protective antibodies and vaccines. Here we report the characterization of a potent neutralizing antibody (N3-1) identified from a COVID-19 patient during the first disease wave. Cryogenic electron microscopy revealed a quaternary binding mode that enables direct interactions with all three receptor-binding domains of the spike protein trimer, resulting in extraordinary avidity and potent neutralization of all major variants of concern until the emergence of Omicron. Structure-based rational design of N3-1 mutants improved binding to all Omicron variants but only partially restored neutralization of the conformationally distinct Omicron BA.1. This study provides new insights into immune evasion through changes in spike protein dynamics and highlights considerations for future conformationally biased multivalent vaccine designs.

A single-dose of intranasal vaccination with a live-attenuated SARS-CoV-2 vaccine candidate promotes protective mucosal and systemic immunity.

An attenuated SARS-CoV-2 virus with modified viral transcriptional regulatory sequences and deletion of open-reading frames 3, 6, 7 and 8 (∆3678) was previously reported to protect hamsters from SARS-CoV-2 infection and transmission. Here we report that a single-dose intranasal vaccination of ∆3678 protects K18-hACE2 mice from wild-type or variant SARS-CoV-2 challenge. Compared with wild-type virus infection, the ∆3678 vaccination induces equivalent or higher levels of lung and systemic T cell, B cell, IgA, and IgG responses. The results suggest ∆3678 as an attractive mucosal vaccine candidate to boost pulmonary immunity against SARS-CoV-2.

Less neutralization evasion of SARS-CoV-2 BA.2.86 than XBB sublineages and CH.1.1.

The highly mutated BA.2.86, with over 30 spike protein mutations in comparison to Omicron BA.2 and XBB.1.5 variants, has raised concerns about its potential to evade COVID-19 vaccination or prior SARS-CoV-2 infection-elicited immunity. In this study, we employ a live SARS-CoV-2 neutralization assay to compare the neutralization evasion ability of BA.2.86 with other emerged SARS-CoV-2 subvariants, including BA.2-derived CH.1.1, Delta-Omicron recombinant XBC.1.6, and XBB descendants XBB.1.5, XBB.1.16, XBB.2.3, EG.5.1 and FL.1.5.1. Our results show that BA.2.86 is less neutralization evasive than XBB sublineages. XBB descendants XBB.1.16, EG.5.1, and FL.1.5.1 continue to significantly evade neutralization induced by the parental COVID-19 mRNA vaccine and a BA.5 Bivalent booster. Notably, when compared to XBB.1.5, the more recent XBB descendants, particularly EG.5.1, display increased resistance to neutralization. Among all the tested variants, CH.1.1 exhibits the greatest neutralization evasion. In contrast, XBC.1.6 shows a slight reduction but remains comparably sensitive to neutralization when compared to BA.5. Furthermore, a recent XBB.1.5-breakthrough infection significantly enhances the breadth and potency of cross-neutralization. These findings reinforce the expectation that the upcoming XBB.1.5 mRNA vaccine would likely boost the neutralization of currently circulating variants, while also underscoring the critical importance of ongoing surveillance to monitor the evolution and immune evasion potential of SARS-CoV-2 variants.

An Integrated Research-Clinical BSL-2 Platform for a Live SARS-CoV-2 Neutralization Assay.

A reliable and efficient serological test is crucial for monitoring neutralizing antibodies against SARS-CoV-2 and its variants of concern (VOCs). Here, we present an integrated research-clinical platform for a live SARS-CoV-2 neutralization assay, utilizing highly attenuated SARS-CoV-2 (Δ3678_WA1-spike). This strain contains mutations in viral transcription regulation sequences and deletion in the open-reading-frames 3, 6, 7, and 8, allowing for safe handling in biosafety level 2 (BSL-2) laboratories. Building on this backbone, we constructed a genetically stable reporter virus (mGFP Δ3678_WA1-spike) by incorporating a modified green fluorescent protein sequence (mGFP). We also constructed mGFP Δ3678_BA.5-spike and mGFP Δ3678_XBB.1.5-spike by substituting the WA1 spike with variants BA.5 and XBB.1.5 spike, respectively. All three viruses exhibit robust fluorescent signals in infected cells and neutralization titers in an optimized fluorescence reduction neutralization assay that highly correlates with a conventional plaque reduction assay. Furthermore, we established that a streamlined robot-aided Bench-to-Clinics COVID-19 Neutralization Test workflow demonstrated remarkably sensitive, specific, reproducible, and accurate characteristics, allowing the assessment of neutralization titers against SARS-CoV-2 variants within 24 h after sample receiving. Overall, our innovative approach provides a valuable avenue for large-scale testing of clinical samples against SARS-CoV-2 and VOCs at BSL-2, supporting pandemic preparedness and response strategies.

Multi-task prediction-based graph contrastive learning for inferring the relationship among lncRNAs, miRNAs and diseases.

MOTIVATION: Identifying the relationships among long non-coding RNAs (lncRNAs), microRNAs (miRNAs) and diseases is highly valuable for diagnosing, preventing, treating and prognosing diseases. The development of effective computational prediction methods can reduce experimental costs. While numerous methods have been proposed, they often to treat the prediction of lncRNA-disease associations (LDAs), miRNA-disease associations (MDAs) and lncRNA-miRNA interactions (LMIs) as separate task. Models capable of predicting all three relationships simultaneously remain relatively scarce. Our aim is to perform multi-task predictions, which not only construct a unified framework, but also facilitate mutual complementarity of information among lncRNAs, miRNAs and diseases. RESULTS: In this work, we propose a novel unsupervised embedding method called graph contrastive learning for multi-task prediction (GCLMTP). Our approach aims to predict LDAs, MDAs and LMIs by simultaneously extracting embedding representations of lncRNAs, miRNAs and diseases. To achieve this, we first construct a triple-layer lncRNA-miRNA-disease heterogeneous graph (LMDHG) that integrates the complex relationships between these entities based on their similarities and correlations. Next, we employ an unsupervised embedding model based on graph contrastive learning to extract potential topological feature of lncRNAs, miRNAs and diseases from the LMDHG. The graph contrastive learning leverages graph convolutional network architectures to maximize the mutual information between patch representations and corresponding high-level summaries of the LMDHG. Subsequently, for the three prediction tasks, multiple classifiers are explored to predict LDA, MDA and LMI scores. Comprehensive experiments are conducted on two datasets (from older and newer versions of the database, respectively). The results show that GCLMTP outperforms other state-of-the-art methods for the disease-related lncRNA and miRNA prediction tasks. Additionally, case studies on two datasets further demonstrate the ability of GCLMTP to accurately discover new associations. To ensure reproducibility of this work, we have made the datasets and source code publicly available at https://github.com/sheng-n/GCLMTP.

PromGER: Promoter Prediction Based on Graph Embedding and Ensemble Learning for Eukaryotic Sequence.

Promoters are DNA non-coding regions around the transcription start site and are responsible for regulating the gene transcription process. Due to their key role in gene function and transcriptional activity, the prediction of promoter sequences and their core elements accurately is a crucial research area in bioinformatics. At present, models based on machine learning and deep learning have been developed for promoter prediction. However, these models cannot mine the deeper biological information of promoter sequences and consider the complex relationship among promoter sequences. In this work, we propose a novel prediction model called PromGER to predict eukaryotic promoter sequences. For a promoter sequence, firstly, PromGER utilizes four types of feature-encoding methods to extract local information within promoter sequences. Secondly, according to the potential relationships among promoter sequences, the whole promoter sequences are constructed as a graph. Furthermore, three different scales of graph-embedding methods are applied for obtaining the global feature information more comprehensively in the graph. Finally, combining local features with global features of sequences, PromGER analyzes and predicts promoter sequences through a tree-based ensemble-learning framework. Compared with seven existing methods, PromGER improved the average specificity of 13%, accuracy of 10%, Matthew's correlation coefficient of 16%, precision of 4%, F1 score of 6%, and AUC of 9%. Specifically, this study interpreted the PromGER by the t-distributed stochastic neighbor embedding (t-SNE) method and SHAPley Additive exPlanations (SHAP) value analysis, which demonstrates the interpretability of the model.

Metformin restrains ZIKV replication and alleviates virus-induced inflammatory responses in microglia.

The re-emergence of Zika virus (ZIKV) remains a major public health threat that has raised worldwide attention. Accumulating evidence suggests that ZIKV can cause serious pathological changes to the human nervous system, including microcephaly in newborns. Recent studies suggest that metformin, an established treatment for diabetes may play a role in viral infection; however, little is known about the interactions between ZIKV infection and metformin administration. Using fluorescent ZIKV by flow cytometry and immunofluorescence imaging, we found that ZIKV can infect microglia in a dose-dependent manner. Metformin diminished ZIKV replication without the alteration of viral entry and phagocytosis. Our study demonstrated that metformin downregulated ZIKV-induced inflammatory response in microglia in a time- and dose-dependent manner. Our RNA-Seq and qRT-PCR analysis found that type I and III interferons (IFN), such as IFNα2, IFNß1 and IFNλ3 were upregulated in ZIKV-infected cells by metformin treatment, accompanied with the downregulation of GBP4, OAS1, MX1 and ISG15. Together, our results suggest that metformin-mediated modulation in multiple pathways may attribute to restraining ZIKV infection in microglia, which may provide a potential tool to consider for use in unique clinical circumstances.

A single-dose of intranasal vaccination with a live-attenuated SARS-CoV-2 vaccine candidate promotes protective mucosal and systemic immunity.

An attenuated SARS-CoV-2 virus with modified viral transcriptional regulatory sequences and deletion of open-reading frames 3, 6, 7 and 8 (∆3678) was previously reported to protect hamsters from SARS-CoV-2 infection and transmission. Here we report that a single-dose intranasal vaccination of ∆3678 protects K18-hACE2 mice from wild-type or variant SARS-CoV-2 challenge. Compared with wild-type virus infection, the ∆3678 vaccination induces equivalent or higher levels of lung and systemic T cell, B cell, IgA, and IgG responses. The results suggest ∆3678 as an attractive mucosal vaccine candidate to boost pulmonary immunity against SARS-CoV-2.

Prediction of Proteins in Cerebrospinal Fluid and Application to Glioma Biomarker Identification.

Cerebrospinal fluid (CSF) proteins are very important because they can serve as biomarkers for central nervous system diseases. Although many CSF proteins have been identified with wet experiments, the identification of CSF proteins is still a challenge. In this paper, we propose a novel method to predict proteins in CSF based on protein features. A two-stage feature-selection method is employed to remove irrelevant features and redundant features. The deep neural network and bagging method are used to construct the model for the prediction of CSF proteins. The experiment results on the independent testing dataset demonstrate that our method performs better than other methods in the prediction of CSF proteins. Furthermore, our method is also applied to the identification of glioma biomarkers. A differentially expressed gene analysis is performed on the glioma data. After combining the analysis results with the prediction results of our model, the biomarkers of glioma are identified successfully.