icumonitoring.ch: a platform for short-term forecasting of intensive care unit occupancy during the COVID-19 epidemic in Switzerland.

In Switzerland, the COVID-19 epidemic is progressively slowing down owing to “social distancing” measures introduced by the Federal Council on 16 March 2020. However, the gradual ease of these measures may initiate a second epidemic wave, the length and intensity of which are difficult to anticipate. In this context, hospitals must prepare for a potential increase in intensive care unit (ICU) admissions of patients with acute respiratory distress syndrome. Here, we introduce icumonitoring.ch, a platform providing hospital-level projections for ICU occupancy. We combined current data on the number of beds and ventilators with canton-level projections of COVID-19 cases from two S-E-I-R models. We disaggregated epidemic projection in each hospital in Switzerland for the number of COVID-19 cases, hospitalisations, hospitalisations in ICU, and ventilators in use. The platform is updated every 3-4 days and can incorporate projections from other modelling teams to inform decision makers with a range of epidemic scenarios for future hospital occupancy.

Évaluation des performances diagnostiques d'infection à SARS-CoV-2 d'un algorithme utilisant l'analyse de signaux physiologiques à partir d'une montre connectée

Introduction Le diagnostic de l'infection SARS-CoV-2 est basé sur la détection du virus par rt-PCR. Une application à été élaborée qui permet d'aider au diagnostic de la COVID-19 sans prélèvement biologique en utilisant des signaux physiologiques recueillis par une montre connectée et un bref questionnaire sur smartphone. L'intelligence artificielle par deep learning/réseaux de neurones a été utilisée pour produire un algorithme de classification diagnostique de COVID-19. Cette étude a pour but d'évaluer les performance diagnostiques de l'application dans des conditions du diagnostic clinique chez des malades et des soignants français. Matériels et méthodes Etude observationnelle multicentrique portant sur des patients hospitalisés et des soignants chez lesquels un test PCR à la recherche de SARS-CoV-2 doit être réalisé (cas suspect ou contact). L'enregistrement se faisait via une montre jumelée à un smartphone dans un délai de 3 jours du test PCR. Recueil d'information clinique et enregistrement par montre connectée de la fréquence cardiaque, de l'intervalle RR, la conductance galvanique de la peau. Les flux de données étaient synchronisés et fenêtrés. L'analyse des résultats porte sur les sensibilité, spécificité, taux de faux positifs et de faux négatifs de l'application par rapport à la PCR. Un questionnaire téléphonique à 7 et 15 jours a été réalisé pour connaitre l'évolution clinique et les résultats d'éventuelles PCR de contrôle. Résultats D'aout 2020 à Mars 2021, 363 participants (239 patients et 94 soignants) ont été inclus et 305 enregistrements étaient analysables. Le jour de l'enregistrement 167 participants étaient asymptomatiques (46 %). Le suivi complet des participants à J7 et J14 a été réalisé pour 248 participants,162 (65.3 %) avait une PCR- et absence de symptôme (P-S-), neuf (3.6 %) des symptomes et PCR neg (P-S+), 62 (25,4 %) un COVID symptomatique PCR+ (P+S+), et 14 (5.6 %) une PCR+ asymptomatique (P+S-). Les données acquises ont été répartie dans différents set d'analyse. Un Training set (52), un Validation set (17) et un Test set (18). Trois modèles de réseaux de neurones ont été entrainé sur les données acquises. Les capacités diagnostiques de l'application ont été évaluées à partir de deux combinaisons de paramètres acquis par la montre. La classification était correcte avec une efficacité (eff) de 99.1 %,un taux de faux positif (FP) de 3 % et pas de faux négatif (FN) pour la première combinaison et eff=97 %, .FP=0, FN=5 % pour la deuxième. Conclusion Les performance diagnostiques de l'application s'appuyant sur des réseaux de neurones entrainés en condition diagnostique chez des malades et des soignants français imitent très bien les performances des tests PCR. L'utilisation d'outils connectés capturant des signaux physiologiques associés à des analyse d'intelligence artificielle et de réseaux de neurones pourrait permettre de réduire le recours aux tests biologiques en ciblant les patients les plus suscpet d'infdection virale active. Liens d'intérêts déclarés Neu Tiger en tant qu'investigateur

Assessing the Mobility of Severe Acute Respiratory Syndrome Coronavirus-2 Spike Protein Glycans by Structural and Computational Methods.

Two years after its emergence, the coronavirus disease-2019 (COVID-19) pandemic caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) remains difficult to control despite the availability of several vaccines. The extensively glycosylated SARS-CoV-2 spike (S) protein, which mediates host cell entry by binding to the angiotensin converting enzyme 2 (ACE2) through its receptor binding domain (RBD), is the major target of neutralizing antibodies. Like to many other viral fusion proteins, the SARS-CoV-2 spike protein utilizes a glycan shield to thwart the host immune response. To grasp the influence of chemical signatures on carbohydrate mobility and reconcile the cryo-EM density of specific glycans we combined our cryo-EM map of the S ectodomain to 4.1 Å resolution, reconstructed from a limited number of particles, and all-atom molecular dynamics simulations. Chemical modifications modeled on representative glycans (defucosylation, sialylation and addition of terminal LacNAc units) show no significant influence on either protein shielding or glycan flexibility. By estimating at selected sites the local correlation between the full density map and atomic model-based maps derived from molecular dynamics simulations, we provide insight into the geometries of the α-Man-(1→3)-[α-Man-(1→6)-]-ß-Man-(1→4)-ß-GlcNAc(1→4)-ß-GlcNAc core common to all N-glycosylation sites.

CovidDeep: SARS-CoV-2/COVID-19 Test Based on Wearable Medical Sensors and Efficient Neural Networks

The novel coronavirus (SARS-CoV-2) has led to a pandemic. The current testing regime based on Reverse Transcription-Polymerase Chain Reaction for SARS-CoV-2 has been unable to keep up with testing demands, and also suffers from a relatively low positive detection rate in the early stages of the resultant COVID-19 disease. Hence, there is a need for an alternative approach for repeated large-scale testing of SARS-CoV-2/COVID-19. The emergence of wearable medical sensors (WMSs) and deep neural networks (DNNs) points to a promising approach to address this challenge. WMSs enable continuous and user-transparent monitoring of physiological signals. However, disease detection based on WMSs/DNNs and their deployment on resource-constrained edge devices remain challenging problems. To address these problems, we propose a framework called CovidDeep that combines efficient DNNs with commercially available WMSs for pervasive testing of the virus and the resultant disease. CovidDeep does not depend on manual feature extraction. It directly operates on WMS data and some easy-to-answer questions in a questionnaire whose answers can be obtained through a smartphone application. We collected data from 87 individuals, spanning three cohorts including healthy, asymptomatic (to detect the virus), and symptomatic (to detect the disease) patients. We trained DNNs on various subsets of the features automatically extracted from six WMS and questionnaire categories to perform ablation studies to determine which subsets are most efficacious in terms of test accuracy for a three-way classification. The highest test accuracy obtained was 98.1%. The models were also shown to perform well on other performance measures, such as false positive rate, false negative rate, and F1 score. We augmented the real training dataset with a synthetic training dataset drawn from the same probability distribution to impose a prior on DNN weights and leveraged a grow-and-prune synthesis paradigm to learn both DNN architecture and weights. This boosted the accuracy of the various DNNs further and simultaneously reduced their size and floating-point operations. This makes the CovidDeep DNNs both accurate and efficient, in terms of memory requirements and computations. The resultant DNNs are embedded in a smartphone application, which has the added benefit of preserving patient privacy. Author

Association of Ocular Adverse Events With Inactivated COVID-19 Vaccination in Patients in Abu Dhabi.

Importance: As vaccinations against COVID-19 continue, potential ocular adverse events should be reported in detail to increase awareness among the medical community, although typically, a causal relationship cannot be established definitively. Objective: To describe ocular adverse events that occur soon after receiving an inactivated COVID-19 vaccination (Sinopharm). Design, Setting, and Participants: This case series took place from September 2020 to January 2021 at Cleveland Clinic Abu Dhabi, a tertiary referral center. Patients who reported ocular adverse events and presented within 15 days from the first of 2 doses of an inactivated COVID-19 vaccine were analyzed. Main Outcomes and Measures: Each patient underwent Snellen best-corrected visual acuity that was then converted to logMAR, applanation tonometry, and biomicroscopic examination with indirect ophthalmoscopy. Color fundus photography was obtained with a conventional 9-field fundus photography camera or with a widefield fundus photography system. Optical coherence tomography and optical coherence tomographic angiography images were obtained. Sex, race, age, and clinical data were self-reported. Results: Nine eyes of 7 patients (3 male individuals) presenting with ocular complaints following COVID-19 vaccine were included in the study. The mean (SD) age was 41.4 (9.3) years (range, 30-55 years); the mean best-corrected visual acuity was 0.23 logMAR (range, 0-1 logMAR; approximate Snellen equivalent, 20/32). The mean time of ocular adverse event manifestations was 5.2 days (range, 1-10 days). One patient was diagnosed with episcleritis, 2 with anterior scleritis, 2 with acute macular neuroretinopathy, 1 with paracentral acute middle maculopathy, and 1 with subretinal fluid. Conclusions and Relevance: In this case series study of 7 patients, the timing of transient and ocular complications 5.2 days after vaccination with an inactivated COVID-19 vaccine supported an association with the ocular findings, but a causal relationship cannot be established from this study design.

Socioeconomic position and the COVID-19 care cascade from testing to mortality in Switzerland: a population-based analysis.

BACKGROUND: The inverse care law states that disadvantaged populations need more health care than advantaged populations but receive less. Gaps in COVID-19-related health care and infection control are not well understood. We aimed to examine inequalities in health in the care cascade from testing for SARS-CoV-2 to COVID-19-related hospitalisation, intensive care unit (ICU) admission, and death in Switzerland, a wealthy country strongly affected by the pandemic. METHODS: We analysed surveillance data reported to the Swiss Federal Office of Public Health from March 1, 2020, to April 16, 2021, and 2018 population data. We geocoded residential addresses of notifications to identify the Swiss neighbourhood index of socioeconomic position (Swiss-SEP). The index describes 1·27 million small neighbourhoods of approximately 50 households each on the basis of rent per m2, education and occupation of household heads, and crowding. We used negative binomial regression models to calculate incidence rate ratios (IRRs) with 95% credible intervals (CrIs) of the association between ten groups of the Swiss-SEP index defined by deciles (1=lowest, 10=highest) and outcomes. Models were adjusted for sex, age, canton, and wave of the epidemic (before or after June 8, 2020). We used three different denominators: the general population, the number of tests, and the number of positive tests. FINDINGS: Analyses were based on 4 129 636 tests, 609 782 positive tests, 26 143 hospitalisations, 2432 ICU admissions, 9383 deaths, and 8 221 406 residents. Comparing the highest with the lowest Swiss-SEP group and using the general population as the denominator, more tests were done among people living in neighbourhoods of highest SEP compared with lowest SEP (adjusted IRR 1·18 [95% CrI 1·02-1·36]). Among tested people, test positivity was lower (0·75 [0·69-0·81]) in neighbourhoods of highest SEP than of lowest SEP. Among people testing positive, the adjusted IRR was 0·68 (0·62-0·74) for hospitalisation, was 0·54 (0·43-0·70) for ICU admission, and 0·86 (0·76-0·99) for death. The associations between neighbourhood SEP and outcomes were stronger in younger age groups and we found heterogeneity between areas. INTERPRETATION: The inverse care law and socioeconomic inequalities were evident in Switzerland during the COVID-19 epidemic. People living in neighbourhoods of low SEP were less likely to be tested but more likely to test positive, be admitted to hospital, or die, compared with those in areas of high SEP. It is essential to continue to monitor testing for SARS-CoV-2, access and uptake of COVID-19 vaccination and outcomes of COVID-19. Governments and health-care systems should address this pandemic of inequality by taking measures to reduce health inequalities in response to the SARS-CoV-2 pandemic. FUNDING: Swiss Federal Office of Public Health, Swiss National Science Foundation, EU Horizon 2020, Branco Weiss Foundation.

Sensitive detection of SARS-CoV-2 seroconversion by flow cytometry reveals the presence of nucleoprotein-reactive antibodies in unexposed individuals.

There is an ongoing need of developing sensitive and specific methods for the determination of SARS-CoV-2 seroconversion. For this purpose, we have developed a multiplexed flow cytometric bead array (C19BA) that allows the identification of IgG and IgM antibodies against three immunogenic proteins simultaneously: the spike receptor-binding domain (RBD), the spike protein subunit 1 (S1) and the nucleoprotein (N). Using different cohorts of samples collected before and after the pandemic, we show that this assay is more sensitive than ELISAs performed in our laboratory. The combination of three viral antigens allows for the interrogation of full seroconversion. Importantly, we have detected N-reactive antibodies in COVID-19-negative individuals. Here we present an immunoassay that can be easily implemented and has superior potential to detect low antibody titers compared to current gold standard serology methods.

Superimposition of Viral Protein Structures: A Means to Decipher the Phylogenies of Viruses.

Superimposition of protein structures is key in unravelling structural homology across proteins whose sequence similarity is lost. Structural comparison provides insights into protein function and evolution. Here, we review some of the original findings and thoughts that have led to the current established structure-based phylogeny of viruses: starting from the original observation that the major capsid proteins of plant and animal viruses possess similar folds, to the idea that each virus has an innate "self". This latter idea fueled the conceptualization of the PRD1-adenovirus lineage whose members possess a major capsid protein (innate "self") with a double jelly roll fold. Based on this approach, long-range viral evolutionary relationships can be detected allowing the virosphere to be classified in four structure-based lineages. However, this process is not without its challenges or limitations. As an example of these hurdles, we finally touch on the difficulty of establishing structural "self" traits for enveloped viruses showcasing the coronaviruses but also the power of structure-based analysis in the understanding of emerging viruses.

Structural Characterization of N-Linked Glycans in the Receptor Binding Domain of the SARS-CoV-2 Spike Protein and their Interactions with Human Lectins.

The glycan structures of the receptor binding domain of the SARS-CoV2 spike glycoprotein expressed in human HEK293F cells have been studied by using NMR. The different possible interacting epitopes have been deeply analysed and characterized, providing evidence of the presence of glycan structures not found in previous MS-based analyses. The interaction of the RBD 13 C-labelled glycans with different human lectins, which are expressed in different organs and tissues that may be affected during the infection process, has also been evaluated by NMR. In particular, 15 N-labelled galectins (galectins-3, -7 and -8 N-terminal), Siglecs (Siglec-8, Siglec-10), and C-type lectins (DC-SIGN, MGL) have been employed. Complementary experiments from the glycoprotein perspective or from the lectin's point of view have permitted to disentangle the specific interacting epitopes in each case. Based on these findings, 3D models of the interacting complexes have been proposed.

Sensitive detection of SARS-CoV-2 seroconversion by flow cytometry reveals the presence of nucleoprotein-reactive antibodies in Covid-19-naive individuals (preprint)

We have developed a novel multiplexed flow cytometric bead array (C19BA) for the detection of SARS-CoV-2 seroconversion that allows sensitive identification of IgG and IgM antibodies against three immunogenic proteins: the spike receptor-binding domain (RBD), the spike protein subunit 1 (S1) and the nucleoprotein (N) simultaneously. This assay is more sensitive than ELISA, and the combination of three antigens allows for the interrogation of full seroconversion. Importantly, we have detected N-reactive antibodies in COVID-19-negative individuals.