Lest We Forget: Covid-19, the Defense Production Act, and Executive Order 13,917

SARS-CoV-2's origin should be investigated worldwide for pandemic prevention

You Can't Make Me Stay Home! Medical and Legal Aspects of the COVID-19 Pandemic

Management of bacteriological specimens of patients suffering from coronavirus disease 2019 (COVID-19)

Utilization of deceased donors during a pandemic: argument against using SARS-CoV-2-positive donors.

Coronavirus Disease 2019 and Vaccination of Children and Adolescents: Prospects and Challenges

COVID-19: What do we know?

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infecting Pregnant Women and the Fetus, Intrauterine Transmission, and Placental Pathology During the Coronavirus Disease 2019 (COVID-19) Pandemic: It's Complicated.

Even though extreme containment and mitigation strategies were implemented by numerous governments around the world to slow down the spread of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), the number of critically ill patients and fatalities keeps rising. This crisis has highlighted the socioeconomic disparities of health care systems within and among countries. As new CoVID policies and responses are implemented to lessen the impact of the virus, it is imperative (1) to consider additional mitigation strategies critical for the development of effective countermeasures, (2) to promote long-term policies and strict regulations of the trade of wildlife and live animal markets, and (3) to advocate for necessary funding and investments in global health, specifically for the prevention of and response to natural and manmade pandemics. This document considers some of these challenges.

Decision letter: COVID-19 pandemic dynamics in South Africa and epidemiological characteristics of three variants of concern (Beta, Delta, and Omicron)

Current COVID-19 guidelines for respiratory protection of health care workers are inadequate.

Editorial: Coronavirus disease 2019 (COVID-19) - advances in epidemiology, diagnostics, treatments, host-directed therapies, pathogenesis, vaccines, and ongoing challenges.

Immunization against Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) is a major step in protecting the healthcare worker beyond standard pandemic precautions and infection control measures. Healthcare workers (HCW) are at high risk for SARS-CoV-2 exposure and infection placing burden on the workplace. Hence, reducing SARS-CoV-2 infection of HCWs serves multiple goals in the occupational health arena. These include protecting the health and safety of the HCW as well as reducing transmission of Coronavirus Disease 2019 (COVID-19) in the clinic and the larger workplace it serves. Furthermore, when HCWs demonstrate appropriate pandemic precautions and health practices, they serve as strategic models for other workers. Here, this modeling represents leadership which can inspire optimism and hope in a workforce beleaguered by the pandemic.

Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Although France was one of the most affected European countries by the COVID-19 pandemic in 2020, the dynamics of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) movement within France, but also involving France in Europe and in the world, remain only partially characterized in this timeframe. Here, we analyzed GISAID deposited sequences from January 1 to December 31, 2020 (n = 638,706 sequences at the time of writing). To tackle the challenging number of sequences without the bias of analyzing a single subsample of sequences, we produced 100 subsamples of sequences and related phylogenetic trees from the whole dataset for different geographic scales (worldwide, European countries, and French administrative regions) and time periods (from January 1 to July 25, 2020, and from July 26 to December 31, 2020). We applied a maximum likelihood discrete trait phylogeographic method to date exchange events (i.e., a transition from one location to another one), to estimate the geographic spread of SARS-CoV-2 transmissions and lineages into, from and within France, Europe, and the world. The results unraveled two different patterns of exchange events between the first and second half of 2020. Throughout the year, Europe was systematically associated with most of the intercontinental exchanges. SARS-CoV-2 was mainly introduced into France from North America and Europe (mostly by Italy, Spain, the United Kingdom, Belgium, and Germany) during the first European epidemic wave. During the second wave, exchange events were limited to neighboring countries without strong intercontinental movement, but Russia widely exported the virus into Europe during the summer of 2020. France mostly exported B.1 and B.1.160 lineages, respectively, during the first and second European epidemic waves. At the level of French administrative regions, the Paris area was the main exporter during the first wave. But, for the second epidemic wave, it equally contributed to virus spread with Lyon area, the second most populated urban area after Paris in France. The main circulating lineages were similarly distributed among the French regions. To conclude, by enabling the inclusion of tens of thousands of viral sequences, this original phylodynamic method enabled us to robustly describe SARS-CoV-2 geographic spread through France, Europe, and worldwide in 2020. Editor's evaluation This paper is a comprehensive, quantitative, and robust overview of the global, European, and French genomic epidemiology of SARS-CoV-2 in the first year of the pandemic. It contributes methodological advances in maximum likelihood phylogeography, using multiple scales and providing a simulation-based validation. The results show two distinct patterns of SARS-CoV-2 exchange events between the first and second half of 2020, with Europe being involved in most intercontinental exchanges: France experienced viral introductions primarily from North America and Europe during the first wave, while the second wave saw limited intercontinental movement and a significant contribution of the virus from Russia into Europe. https://doi.org/10.7554/eLife.82538.sa0 Decision letter Reviews on Sciety eLife's review process Introduction On December 1, 2019, an outbreak of severe respiratory disease was identified in the city of Wuhan, China (Huang et al., 2020). The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was rapidly identified as the agent of the disease (Zhu et al., 2020), responsible for the ongoing global pandemic of coronavirus disease 2019 (COVID-19). By the end of 2020, the virus caused over 1.8 million deaths worldwide including ~65,000 deaths in France, concomitantly with social and economic devastations in many regions of the world (Mofijur et al., 2021; Santomauro et al., 2021). Since the beginning of COVID-19 pandemic, the scientific community has thoroughly characterized the virus, including its pathogenesis, the monitoring of its circulation in human populations, and the development of several treatments or vaccines (Cevik et al., 2020; Krammer, 2020). Epidemiological models have been particularly helpful to quantify viral spread both in the short and long terms and to inform public health decisions (Hoertel et al., 2020; Kissler et al., 2020). In addition to clinical and epidemiological insights, viral whole-genome sequencing has become a powerful and invaluable tool to better understand infection dynamics (Volz et al., 2013), including the COVID-19 pandemic. The number of available SARS-CoV-2 whole-genome sequences has rapidly grown thanks to the efforts of scientists and researchers gathered via international networks such as the Global Initiative on Sharing All Influenza Data, GISAID (https://www.gisaid.org/; Khare et al., 2021). These genomic sequences are essential to effectively reconstruct the global viral spread and the origins of variants. Genomic data have become a strong asset in addition to epidemiological data to inform governments and help public health decisions (Attwood et al., 2022; Rife et al., 2017). However, due to the computational time required for many analyses, existing phylogenetic tools are limited for studying large amounts of data such as those generated by widespread viral sequencing. Therefore, it is still necessary to develop methods to analyze large datasets while optimizing computational calculation times. Producing appropriate subsamples through several replicates may be an efficient approach in this matter. In France, the first COVID-19 suspected case was identified in late December 2019 (Deslandes et al., 2020), and the first confirmed cases of SARS-CoV-2 infection were detected on January 24, 2020, in individuals who had recently traveled in China (Bernard Stoecklin et al., 2020). COVID-19 cases remained scarce until the end of February, when the national incidence curve of new SARS-CoV-2 infections started to rise (Figure 1). By the end of February, reinforced measures were announced, including social distancing, cessation of passenger flights to France, school closure, and finally, a complete lockdown across the entire country from March 17 to May 10, 2020. The reported daily incidence and numbers of severe cases peaked at the beginning of April 2020 before decreasing steadily until August 2020. However, after the relaxation of social distancing measures in June, a second wave of infections occurred in early September peaking at more than 100,000 positive cases and 1300 confirmed deaths in a single day on November 2, 2020 (Figure 1). After this peak, daily incidence and severe COVID-19 cases gradually diminished down to a number of positive daily cases varying between 2000 and 25,000 at the end of 2020 thanks to a second national lockdown applied between October 29 and December 15, 2020. Epidemiological trends were similar in most European countries except for Russia or Romania, where high rates of SARS-CoV-2-related deaths were reported even in the summer of 2020. Of note, the other continents showed different patterns of virus circulation: compared to Europe, the number of deaths increased about 2 weeks later in North America and remained high throughout 2020; and from early May, Asia and South America were also highly impacted by the pandemic (Figure 1—figure supplement 1). Figure 1 with 2 supplements see all Download asset Open asset Timeline of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-related deaths and stringency index in France, 2020. Key events are indicated on the timeline. Official lockdowns included stay home orders and closure of schools and daycares. Based on SARS-CoV-2-related deaths, the two first French epidemic waves are, respectively, dated from March to July 2020, and September to December 2020. SARS-CoV-2-related deaths are displayed as the daily number of deaths (light blue area) and as the weekly average of daily number of deaths (dark blue curve). The stringency (Oxford) index is a composite measure based on different response indicators including school and workplace closures and travel bans, rescaled to a value from 0 to 100 (100 = strictest) (Hale et al., 2021). Elucidating the SARS-CoV-2 dynamic throughout the various phases of the pandemic is paramount to better anticipate how to limit virus circulation for future viral epidemics (Rife et al., 2017). Here, we analyzed GISAID deposited sequences to elucidate the origins and spread of the virus in France, Europe, and the world from January 1 to December 31, 2020. Through a maximum likelihood discrete trait phylogeographic method, we estimated the main geographical areas that contributed to viral introduction into France and Europe, the countries/continents to which France exported SARS-CoV-2 the most and the contribution of the different French regions to the national circulation of the virus. The main exchanged lineages were also investigated. We looked at the differences in virus circulation during each of the two European epidemic waves of 2020 independently. Given France's central geographic location in Europe and the high proportion of international travelers visiting this country before the pandemic, we aimed to explore the role that France played in SARS-CoV-2 exchanges both in Europe and worldwide. Results Defining appropriate subsamples using simulations From January 1 to December 31, 2020, a total of 638,706 sequences were retained in our study. Inferring a phylogenic tree with such a large number of sequences would require very long calculation times. To overcome this limit, we constructed smaller datasets by randomly choosing subsamples (with replacement) of the sequences. The number of sequences for each country at each week was chosen to be proportional to the number of SARS-CoV-2-related deaths per country and per week with a 2-week shift to account for the time between infection and death. As a proof of concept, we conducted an extensive simulation study to estimate the accuracy of the discrete trait phylogeographic inference for rates of transitions between two distinct locations. First, to evaluate the precision of such inference on a tree of 1000 leaves, we simulated a two-states model with different combinations of transition rates in 50 replicates. Parameters were correctly estimated with limited variability across the 50 replicates. The median parameters across replicates gave a very accurate estimation (Figure 2A). Figure 2 with 4 supplements see all Download asset Open asset Estimating variability in transition rates using simulations. (A) Estimated versus true parameters in the simulation study of the two-states model. The two panels show the two transition rates. For each set of parameters, 50 replicates were conducted. The large red dot is the median of the replicates. The red cross is the true parameter value, on the bisector. (B) Estimated rate of transition in subsampled trees. For each replicate (n = 50), one point is the result of one subsampled smaller phylogenetic tree (from a large phylogenetic tree). The big dot shows the median for each replicate. The horizontal red line is the overall median (of the medians), across replicates. The horizontal dashed gray line is the true rate. Only one of the two rate parameters is shown. (C) Log-median error in parameter estimation as a function of the log number of replicates, when inference is conducted on truly independent replicate evolutionary histories, on a tree of 1000 leaves. The points are the data, the dashed line shows the line of slope '−1' which is the expectation as the replicates are truly independent. (D) Log-median error as a function of log number of subsamples used for the inference done on subsampled phylogenetic trees. The colored points and lines show the inference done on 50 distinct realizations of the evolutionary process on the whole tree. The dashed line is the overall regression line with a slope of −0.7. We tested between 1 and 10 subsampled trees (x-axis). Next, to evaluate how independent parameter estimates are done on randomly subsampled trees of the same larger phylogeny, we inferred parameters on 50 100-leaves trees randomly subsampled from a 10,000-leaves SARS-CoV-2 phylogenetic tree. For each resulting subtree, we conducted inferences on 50 replicates corresponding to 50 realizations of the stochastic process of evolution of the discrete character – as done in the first simulation – on the whole tree of 10,000 leaves. For each replicate, we observed some error on the estimation of the parameter, because one replicate only corresponds to one possible realization of the evolutionary process, although the overall median of inferred parameters across subsampled trees was closer to the true parameter values (Figure 2B). Different estimations of the transition rates conducted on different subsampled trees are not expected to be fully independent because the subtrees partly share the same evolutionary history. Therefore, we estimated the level of independence of these estimations. When several estimates are perfectly independent from one another and are averaged to obtain the final estimate of the quantity of interest, we expect the error in parameter estimation to converge to 0 with a 1/N (N−1) scaling, where N is the number of replicates. This is indeed what we observed when we calculated the error on estimation of the parameter as a function of the chosen number of replicates N in the first set of simulations. Here, the replicates were truly independent replicate realizations of the evolutionary history and inference was conducted on the whole tree of 1000 leaves (Figure 2C). On the contrary, when estimates are perfectly dependent, error on the averaged parameter estimate is expected to not decrease with N. When evaluating the error on parameter estimates across subsamples of the large tree, we expected the scaling of error as a function of number of subsamples N to be intermediate between non-independence (~N0 scaling) and perfect independence (~N−1 scaling). Using the relationship between log(error) as a function of log(N), we estimated a slope of −0.7 (Figure 2D). Thus, inferences conducted on subsamples of the same phylogenetic tree are partly independent. The precise degree of independence is expected to depend on the shape of the phylogenetic tree, but the coefficient was similar when doing the same study on a randomly generated tree instead of the SARS-CoV-2 tree. We finally conducted another round of simulations to evaluate the error on what we considered as exchange between multiple locations when using sparse subsampling. For that, a 1,000,000-leaves tree was simulated with a five-states discrete trait representing geographical units. Then, 100 subsampled 1000-leaves trees from the whole phylogenetic tree were produced and the ancestry for the discrete trait was reconstructed from the leaf data only. We estimated the number of transitions (exchanges) of each type and compared them with the one obtained from the main tree, finding a mean error rate of 2.7% over the 100 subsamples (Figure 2—figure supplement 1). Altogether, these simulations suggested that using subsamples of 1000 sequences from a large dataset and performing partially independent replicates seems to be sufficient to accurately estimate transition events. Description of the datasets and global diversity of SARS-CoV-2 sequences We defined 100 subsamples of sequences proportionally to COVID-19 deaths across geographic locations and time for different geographic scales (worldwide, Europe, and French regions) and time periods (from January 1 to July 25, 2020, and from July 26 to December 31, 2020, respectively, covering the first and second European epidemic waves). We chose the sampling intensity guided by the weekly number of SARS-CoV-2-related deaths reported by public health organizations. Here, the number of SARS-CoV-2-related deaths was used rather than the number of detected cases because the latter was biased due to variable ascertainment rates across countries and time. For example, the larger number of PCR tests conducted in the second epidemic wave could wrongly suggest that the virus circulated much more during the second half of 2020 (Figure 1—figure supplement 2). For each geographic scale and time period, there was a positive correlation between the weekly number of SARS-CoV-2-related deaths and the weekly number of sequences we included for a subsample (Spearman's rank correlation, p < 0.001; r = 0.94 for the lowest correlation). We also confirmed that the number of sequences per territory was, on average, properly temporally distributed within each time period (Figure 2—figure supplement 2). Some countries and French administrative regions were however discarded in the analyses because they were not sufficiently represented in the GISAID database. Overall, a total of 39,288 and 39,755 distinct SARS-CoV-2 sequences were included across the 100 sampled phylogenies for the worldwide dataset, respectively, for the first and the second time periods (Table 1). At the European scale, 26,757 and 27,658 different SARS-CoV-2 sequences covering 11 countries were analyzed across the 100 subsamples (Table 1). Focusing on French administrative regions, sequences available on the GISAID database were very sparse. The Provence-Alpes-Côte d'Azur (PACA, Marseille area) was the only region that highly sequenced SARS-CoV-2 in 2020. Île-de-France (IDF, Paris area), Auvergne-Rhône-Alpes (ARA, Lyon area), Occitanie (OCC, Toulouse and Montpellier area), and Bretagne (BRE, Rennes area) have sequenced much less than PACA, but provided sufficient data to investigate SARS-CoV-2 geographic exchange events in France. The remaining French administrative regions were discarded since too few sequences were available to properly match the number of weekly SARS-CoV-2 deaths (Figure 2—figure supplement 3). We thus considered 2543 unique sequences across the 100 subsamples between January 1 and July 25, 2020, and 3124 unique sequences between July 26 and December 31, 2020 (Table 1). Table 1 Number of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) sequences investigated for each dataset. DatasetGeographiesPeriod investigatedAverage number of sequences sampled in a subsampleTotal number of sequencesWorldAfrica, Asia, Europe, France, North America, Oceania, South AmericaJanuary 1 to July 25, 202084639,288July 26 to December 31, 202077739,755EuropeBelgium, France, Germany, Italy, The Netherlands, Poland, Romania, Russia, Spain, Sweden, United KingdomJanuary 1 to July 25, 202090426,757July 26 to December 31, 202087227,658FranceAuvergne-Rhône-Alpes (ARA), Bretagne (BRE), Île-de-France (IDF), Occitanie (OCC), Provence-Alpes-Côte d'Azur (PACA)January 1 to July 25, 20204162543July 26 to December 31, 20204333124 The genomic diversity of circulating SARS-CoV-2 in the different continents, countries, and French regions was found to be similar (Figure 2—figure supplement 4). Overall, genomes showed high sequence conservation compared to the Wuhan-Hu-1 reference in 2020 (mean and median of ~13 single nucleotide polymorphisms (SNPs) with 95% of the distribution comprised between 4 and 25 SNPs). Which continents exchanged SARS-CoV-2 with Europe and France? Through 100 distinct, dated and ancestrally reconstructed phylogenetic trees, we first studied SARS-CoV-2 exchanges worldwide for each of the time periods studied. Between January 1 and July 25, 2020 (covering the first European epidemic wave), we found that Europe (excluding France) accounted for 57.3% of the total number of exportation events, and was the main source of SARS-CoV-2 exportations toward the other continents in all of the subsamples (Figure 3A–D and Figure 3—figure supplement 1). North America also highly participated in virus exportation during this period time (24.3%). South America and Asia were each associated with 7.1% of the total number of exportation events, consistent with a later circulation of the virus in these continents (Figure 1—figure supplement 1). France was estimated to have contributed of the total exportation events, that France was not the European source of SARS-CoV-2 at the international level between January 1 and July 25, 2020. The exportation events from France were mostly toward Europe to a to North America, South America, and Asia (Figure and Figure 3—figure supplement 2). These events mostly of the B.1 and lineages (Figure North America a large proportion of SARS-CoV-2 from other continents of the introduction by South America and Europe (Figure average of of all SARS-CoV-2 introductions were into France, and from North America and Europe (Figure 3—figure supplement 2). These introductions of the B.1 and lineages (Figure The first introductions into France were detected at the beginning of February, and increased to a before the lockdown from March 2020 (Figure Only South America and Asia were associated with a in SARS-CoV-2 introductions after this because such measures were there and the circulation of the virus remained limited in these regions. Figure with 2 supplements see all Download asset Open asset acute respiratory syndrome coronavirus 2 (SARS-CoV-2) exchange worldwide. events were inferred with 100 subsampled phylogenies between January 1 and July 25, 2020, and between July 26 and December 31, 2020. (A) Number of introduction and exportation events for each subsample and for each and France. (B) SARS-CoV-2 exchange between continents and France during the two time periods investigated. In these of a location to the and with an at the is proportional to the exchange (C) Number of exportation and (D) introduction events per territory over time. The mean number of exchanges over the subsamples and for each week was the of the complete lockdowns in France. of lineages exported from France and introduced into France. with a proportion were into the From July 26 to December 31, 2020 European epidemic wave), we observed exchange events worldwide compared to the first half of 2020. we showed the of analyzing several as there was a large in the total number of exportation or introduction events, in Europe (Figure Europe was, as between January 1 and July 25, 2020, the main source of exchanges with a total of of the exportation events across by North America Asia and South America (Figure 3A–D and Figure 3—figure supplement 1). of the events occurred during the summer period to August 2020), corresponding to the summer in most countries of the world. France accounted for of the exportation events, but they were toward other European countries and overall detected from August to November 2020 (Figure and Figure 3—figure supplement consistent with the SARS-CoV-2 incidence in this period in France (Figure 1—figure supplement 2). The B.1.160 accounted for all the exportation events from France (Figure In a similar SARS-CoV-2 introductions into France mostly from Europe (Figure and were detected at a rate from April 2020, at a but limited rate from 2020, and at a strong level in September and October 2020 (Figure 3—figure supplement 2). These SARS-CoV-2 introductions into France in of B.1.160 B.1 and lineages (Figure the virus spread in We aimed to a more of SARS-CoV-2 exchanges between France and other European countries with the same Here, we only on European countries associated with a high incidence and without due to a of data on GISAID (Table 1). By the of introduction and exportation events between January 1 and July 25, 2020 across the we observed that was the to virus exportation toward other European countries, with an average of of the total number of exportation events. The United Kingdom, France, and also highly participated in virus and of and of the total number of exportation events, (Figure and Figure supplement 1). These are in line with epidemiological data, since was the first country in Europe to be affected by the and France, the United Kingdom, and were the other European countries associated with the number of SARS-CoV-2-related deaths during the first wave (Figure 1—figure supplement 1). The number of all exportation events however after the of lockdowns in the different countries (with the first one in on March (Figure France mostly exported SARS-CoV-2 toward and the United and a less toward and the All of these events occurred before the lockdown in France (Figure and Figure supplement and of the B.1 and lineages (Figure The rate of SARS-CoV-2 exportations from France until the second European epidemic wave, as it was also the case for other European countries except Russia (Figure For all introduction events, the were more the United accounted for a of the total number of events, while Russia, Belgium, Germany, Italy, Spain, France, and the represented between and of the total number of events (Figure In France, a high rate of introduction events was observed in and March before the lockdown and mostly from the United and (Figure supplement 2). These introductions in of the and B.1 lineages (Figure Figure 4 with 2 supplements see all Download asset Open asset acute respiratory syndrome coronavirus 2 (SARS-CoV-2) exchanges on the European events were calculated by the results from 100 subsampled phylogenies between January 1 and July 25, 2020, and between July 26 and December 31, 2020. (A) Number of introduction and exportation events for each subsample and for each European (B) SARS-CoV-2 exchange between European countries during the two time periods investigated. In these of a location to the and with an at is proportional to the exchange (C) Number of exportation and (D) introduction events per territory over time. The mean number of exchanges over subsamples and for each week was the of the complete lockdowns in France. of lineages exported from France and introduced into France. with a proportion were into the The second time period (from July 26 to December 31, showed a different of exchanges. Here, we estimated exchanges compared to the first half of 2020. Russia accounted for most of the exportation events (Figure These events were estimated to during the the relaxation of measures in most European and the summer periods (Figure This result was expected since Russia was the European country to a high number of SARS-CoV-2-related deaths during this period (Figure 1—figure supplement 1). France and the United also highly participated in virus exportation (Figure and Figure supplement 1). of these events were detected between August and October 2020 (Figure and before the second lockdown in most European countries first one in on October 2020). these are consistent with epidemiological as was the first country in European to be associated with a of SARS-CoV-2-related deaths, rapidly by France. France mostly exported the virus toward and (Figure and Figure supplement and mostly the B.1.160 (Figure Focusing on introduction events, accounted for of the total number of by the United and France For the remaining European countries, the proportion of introduction events was comprised between and

Editor's evaluation: Phylodynamics of SARS-CoV-2 in France, Europe, and the world in 2020

Serological identification of past and recent SARS-CoV-2 infection through antibody screening in Luanda, Angola.