On the Impact of Voice Anonymization on Speech-Based COVID-19 Detection (preprint)

With advances seen in deep learning, voice-based applications are burgeoning, ranging from personal assistants, affective computing, to remote disease diagnostics. As the voice contains both linguistic and paralinguistic information (e.g., vocal pitch, intonation, speech rate, loudness), there is growing interest in voice anonymization to preserve speaker privacy and identity. Voice privacy challenges have emerged over the last few years and focus has been placed on removing speaker identity while keeping linguistic content intact. For affective computing and disease monitoring applications, however, the paralinguistic content may be more critical. Unfortunately, the effects that anonymization may have on these systems are still largely unknown. In this paper, we fill this gap and focus on one particular health monitoring application: speech-based COVID-19 diagnosis. We test two popular anonymization methods and their impact on five different state-of-the-art COVID-19 diagnostic systems using three public datasets. We validate the effectiveness of the anonymization methods, compare their computational complexity, and quantify the impact across different testing scenarios for both within- and across-dataset conditions. Lastly, we show the benefits of anonymization as a data augmentation tool to help recover some of the COVID-19 diagnostic accuracy loss seen with anonymized data.

Proteomics Investigation of Diverse Serological Patterns in COVID-19 (preprint)

Serum antibodies IgM and IgG are elevated during COVID-19 to defend against viral attack. Atypical results such as negative and abnormally high antibody expression were frequently observed whereas the underlying molecular mechanisms are elusive. In our cohort of 144 COVID-19 patients, 3.5% were both IgM and IgG negative whereas 29.2% remained only IgM negative. The remaining patients exhibited positive IgM and IgG expression, with 9.3% of them exhibiting over 20-fold higher titers of IgM than the others at their plateau. IgG titers in all of them were significantly boosted after vaccination in the second year. To investigate the underlying molecular mechanisms, we classed the patients into four groups with diverse serological patterns and analyzed their two-year clinical indicators. Additionally, we collected 111 serum samples for TMTpro-based longitudinal proteomic profiling and characterized 1494 proteins in total. We found that the continuously negative IgM and IgG expression during COVID-19 were associated with mild inflammatory reactions and high T cell responses. Low levels of serum IgD, inferior complement 1 activation of complement cascades, and insufficient cellular immune responses might collectively lead to compensatory serological responses, causing overexpression of IgM. Serum CD163 was positively correlated with antibody titers during seroconversion. This study suggests that patients with negative serology still developed cellular immunity for viral defense, and that high titers of IgM might not be favorable to COVID-19 recovery.

SARS-CoV-2 Nsp6 causes cardiac defects through MGA/MAX complex-mediated increased glycolysis (preprint)

SARS-CoV-2 infection causes COVID-19, a severe acute respiratory disease associated with cardiovascular complications including long-term outcomes. The presence of virus in cardiac tissue of patients with COVID-19 suggests this is a direct, rather than secondary, effect of infection. By expressing individual SARS-CoV-2 proteins in the Drosophila heart we demonstrated interaction of virus Nsp6 with host proteins of the MGA/MAX complex (MGA, PCGF6 and TFDP1). Complementing transcriptomic data from the fly heart revealed that this interaction blocks the antagonistic MGA/MAX complex, which shifts the balance towards MYC/MAX and activates glycolysis—with similar findings in mouse cardiomyocytes. Further, the Nsp6-induced glycolysis disrupted cardiac mitochondrial function, known to increase reactive oxygen species (ROS) in heart failure; this could explain COVID-19-associated cardiac pathology. Furthermore, inhibiting the glycolysis pathway by 2-deoxy-D-glucose (2DG) treatment attenuated the Nsp6-induced cardiac phenotype in fly and mice; thus, suggesting glycolysis as a potential pharmacological target for treating COVID-19-associated heart failure.

Pulmonary and Renal Long COVID at Two-year Revisit (preprint)

More than 450 million individuals have recovered from COVID-19, but little is known about the host responses to long COVID. We performed proteomic and metabolomic analyses of 991 blood and urine specimens from 144 COVID-19 patients with comprehensive clinical data and up to 763 days of follow up. Our data showed that the lungs and kidneys are the most vulnerable organs in long COVID patients. Pulmonary and renal long COVID of one-year revisit can be predicted by a machine learning model based on clinical and multi-omics data collected during the first month from the disease onset with an ACC of 87.5%. Serum protein SFTPB and ATR were associated with pulmonary long COVID and might be potential therapeutic targets. Notably, our data show that all the patients with persistent pulmonary ground glass opacity or patchy opacity lesions developed into pulmonary fibrosis at two-year revisit. Together, this study depicts the longitudinal clinical and molecular landscape of COVID-19 with up to two-year follow-up and presents a method to predict pulmonary and renal long COVID.

Experimental study of the dispersion of cough-generated droplets from a person going up- or downstairs

The dispersion of cough-generated droplets from a person going up- or downstairs was investigated through a laboratory experiment in a water tunnel. This experiment was carried out with a manikin mounted at inclination angles facing the incoming flow to mimic a person going up or down. Detailed velocity measurements and flow visualization were conducted in the water tunnel experiments. To investigate the influence of the initial position on the motion of particles, a virtual particle approach was adopted to simulate the dispersion of particles using the measured velocity field. Particle clustering, which is caused by the unsteadiness of the flow, was observed in both flow visualization and virtual particle simulation. For the case of going upstairs, particles are concentrated below the person’s shoulder and move downward with a short travel distance. For the case of going downstairs, particles dispersing over the person’s head advect over for a long distance. We also found that the motion of the particles is closely related to the initial position. According to the results in this study, suggestions for the prevention of respiratory infectious disease are made.

Integration of protein context improves protein- based COVID-19 patient stratification (preprint)

Background: Classification of disease severity is crucial for the management of COVID-19. Several studies have shown that individual proteins can be used to classify the severity of COVID-19. Here, we aimed to investigate whether integrating the four types of protein context data, namely, protein complexes, stoichiometric ratios, pathways and network degrees will improve the severity classification of COVID-19. Methods: A SWATH-based proteomic data set of 54 sera samples from 40 COVID-19 patients was employed as the training cohort. Results: Machine learning prioritized two complexes, one stoichiometric ratio, five pathways, twelve proteins and five network degrees. A model based on these 25 features led to effective classification of severe cases with an AUC of 0.965, outperforming the models with proteins only. Complement component C9, transthyretin (TTR) and TTR-RBP complex, the stoichiometric ratio of SAA2/ YLPM1, and the network extent of SIRT7 and A2M were highlighted in this classifier. This classifier was further validated with a TMT-based proteomic data set from the same cohort and an independent SWATH-based proteomic data set from Germany, reaching an AUC of 0.900 and 0.908, respectively. Machine learning models integrating protein context information achieved higher AUCs than models with only one feature type. Conclusion: Our results show that the integration of protein context including protein complexes, stoichiometric ratios, pathways, network degrees, and proteins improves phenotype prediction.

Feasibility to Reduce Swab RT-PCR Tests by Serum Proteomics for COVID-19 Disease Course Monitoring (preprint)

The diagnosis and disease course monitoring of COVID-19 are mainly based on RT-PCR analysis of RNAs extracted from pharyngeal or nasopharyngeal swabs with potential live virus, posing a high risk to medical practitioners. Here, we investigated the feasibility of applying serum proteomics to classify COVID-19 patients in the nucleic acid positive (NCP) and negative (NCN) stages. We analyzed the proteome of 320 inactivated serum samples from 144 COVID-19 patients, and 45 controls and shortlisted 42 regulated proteins in the severe group and 12 regulated proteins in the non-severe group. Together with several key clinical indexes including days after symptom onset, platelet counts and magnesium, we developed machine learning models to classify NCP and NCN with an AUC of 0.94 for the severe cases and 0.89 for the non-severe cases. This study suggests the feasibility of utilizing quantitative serum proteomics for NCP-NCN classification.Funding: This work was supported by grants from the National Key R&D Program of China(No. 2020YFE0202200), National Natural Science Foundation of China (81672086), Zhejiang Province Analysis Test Project (2018C37032), the National Natural Science Foundation of China (81972492, 21904107), Zhejiang Provincial Natural Science Foundation for Distinguished Young Scholars (LR19C050001), Zhejiang Medical and Health Science and Technology Plan (2021KY394), Hangzhou Agriculture andSociety Advancement Program (20190101A04), and Westlake Education Foundation, Tencent Foundation.Conflict of Interest: Tiannan Guo is shareholder of Westlake Omics Inc. W.G. and N.X. are employees of Westlake Omics Inc. The remaining authors declare no competing interests.Ethical Approval: This study has been approved by both the Ethical/Institutional Review Boards of Taizhou Hospital and Westlake University. Informed contents from patients were waived by the boards.

Proteomic and Metabolomic Investigation of COVID-19 Patients with Elevated Serum Lactate Dehydrogenase (preprint)

Serum lactate dehydrogenase (LDH) has been established as a prognostic indicator given its differential expression in COVID-19 patients. However, the molecular mechanisms underneath remain poorly understood. In this study, 144 COVID-19 patients were enrolled to monitor the clinical and laboratory parameters over three weeks. Serum lactate dehydrogenase (LDH) was shown elevated in the COVID-19 patients on admission and declined during the convalescence period, and its ability to classify patient severity outperformed other clinical indicators. A threshold of 247 U/L serum LDH on admission was determined for severity prognosis. Next, we classified a subset of 14 patients into high- and low-risk groups based on serum LDH expression and compared their quantitative serum proteomic and metabolomic differences. The results found COVID-19 patients with high serum LDH exhibited differentially expressed blood coagulation and immune responses including acute inflammatory responses, platelet degranulation, complement cascade, as well as multiple different metabolic responses including lipid metabolism, protein ubiquitination and pyruvate fermentation. Specifically, activation of hypoxia responses was highlighted in patients with high LDH expressions. Taken together, our data showed that serum LDH levels is associated COVID-19 severity, and that elevated serum LDH might be consequences of hypoxia and tissue injuries induced by inflammation.

Functional Analysis of SARS-CoV-2 Proteins in Drosophila Identifies Orf6-induced Pathogenicity Attenuated by Selinexor (preprint)

Background: SARS-CoV-2 causes COVID-19 with a widely diverse disease profile that affect many different tissues. The mechanisms underlying its pathogenicity in host organisms remain unclear. Animal models for study the pathogenicity of SARS-CoV-2 proteins are lacking. Methods: : Using bioinformatic analysis, we showed that the majority of the virus-host interacting proteins are conserved in Drosophila . Therefore, we generated a series of transgenic lines for individual SARS-CoV-2 genes and used the Gal4-UAS system to express them in various tissues to study their pathogenicity. Results: : We found that the Nsp6, Orf6 and Orf7a transgenic flies displayed reduced trachea branching and muscle deficits resulting in “held-up” wing phenotype and poor climbing ability. Furthermore, muscle tissue in these flies showed dramatically reduced mitochondria. Since Orf6 was found to bind nucleopore proteins XPO1, we tested Selinexor, a drug that inhibits XPO1, and found that it could attenuated the Orf6-induced lethality and tissue-specific phenotypes in flies. Conclusions: : Our studies here established new Drosophila models for studying the function of SARS-CoV2 genes, identified Orf6 as a highly pathogenic protein in various tissues, and demonstrated the effects of Selinexor for inhibiting Orf6 toxicity with an in vivo model system.

Genetic evolution analysis of SARS-CoV-2 ORF 1ab/S/M protein and screening of epitopes of mRNA vaccine

Objective　To analyze the genetic and evolutionary properties of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) ORF 1ab/S/M proteins and select antigen epitope sequences of mRNA vaccines. Methods　We analyzed the worldwide SARS-CoV-2 genome sequences in this study and have focused on the protein and nucleic acid sequences of the ORF 1ab/S/M. The neighbor-joining tree was employed to map the global distribution of genetic differences. Based on current research on SARS-CoV-2 and SARS-CoV-2 genetic differences, we predicted candidate mRNA vaccines for SARS-CoV-2. Results　The SARS-CoV-2 ORF 1ab nucleic acid sequence similarity is 100.0%, while the homology is 99.3% in the global hot region;the S-protein nucleic acid sequence similarity is 100.0%, while the homology is 97.5%;the M-protein nucleic acid sequence similarity is 100.0%, while the homology is 99.9%. Global distribution of ORF 1ab/S/M proteins indicates that there is a significant genetic difference between the Americas and Eurasia. Potential vaccine antigen epitope mRNA sequences (11 B cell responses and 13 T cell responses) were selected for SARS-CoV-2 ORF 1ab protein;6 B cell responses and 4 T cell responses antigen epitope mRNA sequences were selected for the Spike protein;3 B cell responses and 7 T cell responses antigen epitope mRNA sequences were selected for the membrane protein. Conclusion　There are significant genetic differences in the global hot spot of SARS-CoV-2 in the Americas and Eurasia. Through our new antigen design strategy to screen linear epitopes, we predicted many sequences in ORF 1ab/S/M coding region that potentially raising an immune response. Our study will benefit the discovery of the mRNA vaccine (tandem antigen epitope sequence), antibody discovery, and potentially understanding related immune mechanisms.

Establishment and Functioning of a Temporary 1000-Bed COVID-19 Hospital During the Initial Outbreak in Wuhan (preprint)

Background: COVID-19 Emerged as a novel zoonotic disease in late 2019 and quickly spread across Wuhan before spreading to other parts of China and rest of the world. Due to the rapid spread of the disease, local hospitals were inundated with COVID-19 patients putting a strain on the healthcare system. Little was known about the transmission and potential clinical management of COVID-19 at that time.Methods: A temporary COVID-19 hospital was built within one week. The confirmed COVID-19 cases were either directly recruited to the hospital or were transferred from other hospitals. Patients were admitted for both quarantine and treatment, as required. Data were collected as part of standard clinical care and retrospectively analyzed.Findings: A total of 2,959 patients were recruited during the operation period of this hospital between February 4, 2020, and April 8, 2020. These patients included 838 severe patients of which 72 were classified as critical, and 66 patients died. No infection was reported among healthcare workers.Interpretation: Setting up a dedicated hospital for COVID-19 provided a critical resource during the peak of the pandemic in Wuhan by enabling both quarantine and treatment for the infected patients. The mortality in this hospital was comparable to other hospitals at the time. These data suggest that this approach may prove beneficial in controlling infectious disease spread and limit mortality and prevent strain on existing healthcare system to enable them to care for non-COVID-19 patients.Funding Statement: This study was supported by funding from Beijing Nova Program Interdisciplinary Cooperation Project (DC; No. Z191100001119021), Chinese PLA General Hospital Youth Project (DC; No.QNF19074), Beijing Nova Program Project (DC; No. Z171100001117012), and China 13th Five-year National Key Grant (LXX; No.2018ZX09201013).Declaration of Interests: The authors declare that there are no competing interests.Ethics Approval Statement: This study was approved by the ethics committee of the Chinese PLA General Hospital, with a waiver of informed consent.

Eleven Routine Clinical Features Predict COVID-19 Severity Uncovered by Machine Learning of Longitudinal Measurements (preprint)

Background: Severity prediction of COVID-19 remains one of the major clinical challenges for the ongoing pandemic. In this study, we aim to establish a model for COVID-19 severity prediction and depict dynamic changes of key clinical features over 7 weeks.Methods: In our retrospective study, a total of 841 patients have been screened with the SARS-CoV-2 nucleic acid test, of which 144 patients were virus RNA (COVID-19) positive, resulting in a data matrix containing of 3,065 readings for 124 types of measurements from 17 categories. We built a support vector machine model assisted with genetic algorithm for feature selection based on the longitudinal measurement. 25 patients as a test cohort were included from an independent hospital.Findings: A panel of 11 routine clinical factors constructed a classifier for COVID-19 severity prediction, achieving an accuracy of over 94%. Validation of the model in an independent cohort containing 25 patients achieved an accuracy of 80%. The overall sensitivity, specificity, PPV and NPV were 0.70, 0.99, 0.93 and 0.93, respectively. This study presents a practical model for timely severity prediction for COVID-19, which is freely available at a webserver https://guomics.shinyapps.io/covidAI/.Interpretation: The model provided a classifier composed of 11 routine clinical features which are widely available during COVID-19 management which could predict the severity and may guide the medical care of COVID-19 patients.Funding: This work is supported by grants from Tencent Foundation (2020), National Natural Science Foundation of China (81972492, 21904107, 81672086), Zhejiang Provincial Natural Science Foundation for Distinguished Young Scholars (LR19C050001), Hangzhou Agriculture and Society Advancement Program (20190101A04).Declaration of Interests: NAEthics Approval Statement: This study was approved by the Medical Ethics Committee of Taizhou Hospital, Shaoxing People’s Hospital and Westlake University, Zhejiang province of China, and informed consent was obtained from each enrolled subject.

SARS-CoV-2 ORF9c Is a Membrane-Associated Protein that Suppresses Antiviral Responses in Cells (preprint)

Disrupted antiviral immune responses are associated with severe COVID-19, the disease caused by SAR-CoV-2. Here, we show that the 73-amino-acid protein encoded by ORF9c of the viral genome contains a putative transmembrane domain, interacts with membrane proteins in multiple cellular compartments, and impairs antiviral processes in a lung epithelial cell line. Proteomic, interactome, and transcriptomic analyses, combined with bioinformatic analysis, revealed that expression of only this highly unstable small viral protein impaired interferon signaling, antigen presentation, and complement signaling, while inducing IL-6 signaling. Furthermore, we showed that interfering with ORF9c degradation by either proteasome inhibition or inhibition of the ATPase VCP blunted the effects of ORF9c. Our study indicated that ORF9c enables immune evasion and coordinates cellular changes essential for the SARS-CoV-2 life cycle. One-sentence summarySARS-CoV-2 ORF9c is the first human coronavirus protein localized to membrane, suppressing antiviral response, resembling full viral infection.

Identification of potential key genes for SARS-CoV-2 infected human bronchial organoids based on bioinformatics analysis (preprint)

There is an urgent need to understand the pathogenesis of the severe acute respiratory syndrome coronavirus clade 2 (SARS-CoV-2) that leads to COVID-19 and respiratory failure. Our study is to discover differentially expressed genes (DEGs) and biological signaling pathways by using a bioinformatics approach to elucidate their potential pathogenesis. The gene expression profiles of the GSE150819 datasets were originally produced using an Illumina NextSeq 500 (Homo sapiens). KEGG (Kyoto Encyclopedia of Genes and Genomes) and GO (Gene Ontology) were utilized to identify functional categories and significant pathways. KEGG and GO results suggested that the Cytokine-cytokine receptor interaction, P53 signaling pathway, and Apoptosis are the main signaling pathways in SARS-CoV-2 infected human bronchial organoids (hBOs). Furthermore, NFKBIA, C3, and CCL20 may be key genes in SARS-CoV-2 infected hBOs. Therefore, our study provides further insights into the therapy of COVID-19.

Multi-organ Proteomic Landscape of COVID-19 Autopsies (preprint)

The molecular pathology of multi-organ injuries in COVID-19 patients remains unclear, preventing effective therapeutics development. Here, we report an in-depth multi-organ proteomic landscape of COVID-19 patient autopsy samples. By integrative analysis of proteomes of seven organs, namely lung, spleen, liver, heart, kidney, thyroid and testis, we characterized 11,394 proteins, in which 5336 were perturbed in COVID-19 patients compared to controls. Our data showed that CTSL, rather than ACE2, was significantly upregulated in the lung from COVID-19 patients. Dysregulation of protein translation, glucose metabolism, fatty acid metabolism was detected in multiple organs. Our data suggested upon SARS-CoV-2 infection, hyperinflammation might be triggered which in turn induces damage of gas exchange barrier in the lung, leading to hypoxia, angiogenesis, coagulation and fibrosis in the lung, kidney, spleen, liver, heart and thyroid. Evidence for testicular injuries included reduced Leydig cells, suppressed cholesterol biosynthesis and sperm mobility. In summary, this study depicts the multi-organ proteomic landscape of COVID-19 autopsies, and uncovered dysregulated proteins and biological processes, offering novel therapeutic clues. HIGHLIGHTSO_LICharacterization of 5336 regulated proteins out of 11,394 quantified proteins in the lung, spleen, liver, kidney, heart, thyroid and testis autopsies from 19 patients died from COVID-19. C_LIO_LICTSL, rather than ACE2, was significantly upregulated in the lung from COVID-19 patients. C_LIO_LIEvidence for suppression of glucose metabolism in the spleen, liver and kidney; suppression of fatty acid metabolism in the kidney; enhanced fatty acid metabolism in the lung, spleen, liver, heart and thyroid from COVID-19 patients; enhanced protein translation initiation in the lung, liver, renal medulla and thyroid. C_LIO_LITentative model for multi-organ injuries in patients died from COVID-19: SARS-CoV-2 infection triggers hyperinflammatory which in turn induces damage of gas exchange barrier in the lung, leading to hypoxia, angiogenesis, coagulation and fibrosis in the lung, kidney, spleen, liver, heart, kidney and thyroid. C_LIO_LITesticular injuries in COVID-19 patients included reduced Leydig cells, suppressed cholesterol biosynthesis and sperm mobility. C_LI

NVX-CoV2373 vaccine protects cynomolgus macaque upper and lower airways against SARS-CoV-2 challenge (preprint)

There is an urgent need for a safe and protective vaccine to control the global spread of SARS-CoV-2 and prevent COVID-19. Here, we report the immunogenicity and protective efficacy of a SARS-CoV-2 subunit vaccine (NVX-CoV2373) produced from the full-length SARS-CoV-2 spike (S) glycoprotein stabilized in the prefusion conformation. Cynomolgus macaques (Macaca fascicularis) immunized with NVX-CoV2373 and the saponin-based Matrix-M adjuvant induced anti-S antibody that was neutralizing and blocked binding to the human angiotensin-converting enzyme 2 (hACE2) receptor. Following intranasal and intratracheal challenge with SARS-CoV-2, immunized macaques were protected against upper and lower infection and pulmonary disease. These results support ongoing phase 1/2 clinical studies of the safety and immunogenicity of NVX-CoV2327 vaccine (NCT04368988). HighlightsO_LIFull-length SARS-CoV-2 prefusion spike with Matrix-M1 (NVX-CoV2373) vaccine. C_LIO_LIInduced hACE2 receptor blocking and neutralizing antibodies in macaques. C_LIO_LIVaccine protected against SARS-CoV-2 replication in the nose and lungs. C_LIO_LIAbsence of pulmonary pathology in NVX-CoV2373 vaccinated macaques. C_LI

Eleven Routine Clinical Features Predict COVID-19 Severity (preprint)

Severity prediction of COVID-19 remains one of the major clinical challenges for the ongoing pandemic. Here, we have recruited a 144 COVID-19 patient cohort consisting of training, validation, and internal test sets, longitudinally recorded 124 routine clinical and laboratory parameters, and built a machine learning model to predict the disease progression based on measurements from the first 12 days since the disease onset when no patient became severe. A panel of 11 routine clinical factors, including oxygenation index, basophil counts, aspartate aminotransferase, gender, magnesium, gamma glutamyl transpeptidase, platelet counts, activated partial thromboplastin time, oxygen saturation, body temperature and days after symptom onset, constructed a classifier for COVID-19 severity prediction, achieving accuracy of over 94%. Validation of the model in an independent cohort containing 25 patients achieved accuracy of 80%. The overall sensitivity, specificity, PPV and NPV were 0.70, 0.99, 0.93 and 0.93, respectively. Our model captured predictive dynamics of LDH and CK while their levels were in the normal range. This study presents a practical model for timely severity prediction and surveillance for COVID-19, which is freely available at webserver https://guomics.shinyapps.io/covidAI/.

Gut Microbiota May Underlie the Predisposition of Healthy Individuals to COVID-19-Sensitive Proteomic Biomarkers (preprint)

Background: The COVID-19 pandemic is spreading globally with high disparity in the susceptibility of the disease severity. Identification of the key underlying factors for this disparity is highly warranted. Results: Here we describe constructing a proteomic risk score (PRS) based on 20 blood proteomic biomarkers which related to the progression to severe COVID-19. Among COVID-19 patients, per 10% increment in the PRS was associated with a 57% higher risk of progressing to clinically severe phase (RR=1.57; 95% CI, 1.35-1.82). We demonstrate that in our own cohort of 990 individuals without infection, this proteomic risk score is positively associated with proinflammatory cytokines mainly among older, but not younger, individuals. We further discovered that a core set of gut microbiota could accurately predict the blood proteomic biomarkers of COVID-19 using a machine learning model. The core OTU-predicted PRS had a significant correlation with actual PRS both cross-sectionally (n=132, p<0.001) and prospectively (n=169, p<0.05). Most of the core OTUs were highly correlated with proinflammatory cytokines. Fecal metabolomics analysis suggested potential amino acid-related pathways linking the above core gut microbiota to inflammation.Conclusions: Our study suggests that gut microbiota may underlie the predisposition of healthy individuals to COVID-19-sensitive proteomic biomarkers.