key: cord-0996478-du5bdjlp authors: Yadouleton, A.; Sander, A.-L.; Moreira-Soto, A.; Tchibozo, C.; Hounkanrin, G.; Badou, Y.; Fischer, C.; Krause, N.; Akogbeto, P.; de Oliveira Filho, E. F.; Dossou, A.; Bruenink, S.; AIssi, M.; Djingarey, M. H.; Hounkpatin, B.; Nagel, M.; Drexler, J. f. title: Diagnostics and spread of SARS-CoV-2 in Western Africa: An observational laboratory-based study from Benin date: 2020-07-08 journal: nan DOI: 10.1101/2020.06.29.20140749 sha: 3ea2fd629542b567d7c0f5e3756304b3e620f4bd doc_id: 996478 cord_uid: du5bdjlp Information on severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) spread in Africa is limited by fragile 2 surveillance systems and insufficient diagnostic capacity. 3 We assessed the coronavirus disease-19 (COVID-19)-related diagnostic workload in Benin, Western Africa, 4 characterized SARS-CoV-2 genomes from 12 acute cases of COVID-19, used those together with public data to 5 estimate SARS-CoV-2 transmission dynamics in a Bayesian framework, validated a widely used diagnostic dual target 6 RT-PCR kit donated to African countries, and conducted serological analyses in 68 sera from confirmed COVID-19 7 cases and from febrile patients sampled before the predicted SARS-CoV-2 introduction. 8 We found a 15-fold increase in the monthly laboratory workload due to COVID-19. Genomic surveillance showed 9 introductions of three distinct SARS-CoV-2 lineages. SARS-CoV-2 genome-based analyses yielded an R0 estimate of 10 4.4 (95% confidence interval: 2.0-7.7), suggesting intense spread of SARS-CoV-2 in Africa. RT-PCR-based tests 11 were highly sensitive but showed variation of internal controls and between diagnostic targets. Commercially available 12 SARS-CoV-2 ELISAs showed up to 25% false-positive results depending on antigen and antibody types, likely due 13 to unspecific antibody responses elicited by acute malaria according to lack of SARS-CoV-2-specific neutralizing 14 antibody responses and relatively higher parasitemia in those sera. 15 We confirm an overload of the diagnostic capacity in Benin and provide baseline information on the usability of 16 genome-based surveillance in resource-limited settings. Sero-epidemiological studies needed to assess SARS-CoV-2 17 spread may be put at stake by low specificity of tests in tropical settings globally. The increasing diagnostic challenges 18 demand continuous support of national and supranational African stakeholders. 2 Abstract 1 Information on severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) spread in Africa is limited by fragile 2 surveillance systems and insufficient diagnostic capacity. 3 We assessed the coronavirus disease-19 (COVID-19)-related diagnostic workload in Benin, Western Africa, 4 characterized SARS-CoV-2 genomes from 12 acute cases of COVID-19, used those together with public data to 5 estimate SARS-CoV-2 transmission dynamics in a Bayesian framework, validated a widely used diagnostic dual target 6 RT-PCR kit donated to African countries, and conducted serological analyses in 68 sera from confirmed COVID-19 7 cases and from febrile patients sampled before the predicted SARS-CoV-2 introduction. We found a 15-fold increase in the monthly laboratory workload due to COVID-19. Genomic surveillance showed 9 introductions of three distinct SARS-CoV-2 lineages. SARS-CoV-2 genome-based analyses yielded an R0 estimate of 10 4.4 (95% confidence interval: 2.0-7.7), suggesting intense spread of SARS-CoV-2 in Africa. RT-PCR-based tests 11 were highly sensitive but showed variation of internal controls and between diagnostic targets. Commercially available 12 SARS-CoV-2 ELISAs showed up to 25% false-positive results depending on antigen and antibody types, likely due 13 to unspecific antibody responses elicited by acute malaria according to lack of SARS-CoV-2-specific neutralizing 14 antibody responses and relatively higher parasitemia in those sera. We confirm an overload of the diagnostic capacity in Benin and provide baseline information on the usability of 16 genome-based surveillance in resource-limited settings. Sero-epidemiological studies needed to assess SARS-CoV-2 17 spread may be put at stake by low specificity of tests in tropical settings globally. The increasing diagnostic challenges 18 demand continuous support of national and supranational African stakeholders. Coronavirus infectious disease-19 (COVID-19) emerged in China late 2019 and has afforded over 5 million cases 2 globally by early June 2020. The large numbers of cases cause pressure to health care systems worldwide including 3 laboratory diagnostics of the causative severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). Africa may 4 be particularly vulnerable to As shown in Figure 1A , Sub-Saharan Africa is the most underdeveloped 5 region globally according to the United Nations Development Programme (UNDP). Weak health care and surveillance 6 systems may fail to control SARS-CoV-2 spread and provide basic medical attention as evidenced by the low UNDP 7 health indicators of quantity of physicians and hospital beds per 10,000 people ( Figure 1B) . 2 Age is a major risk factor 8 for severe COVID-19 globally, hence the relatively younger African population may be relatively more protected 9 ( Figure 1A) . However, frequent cardiovascular and chronic respiratory diseases, malaria, HIV infection and 10 tuberculosis combined with unhealthy diets may increase risks for severe courses of COVID-19 in Africa despite a 11 relatively young population. 3 To date, over 70 thousand confirmed COVID-19 cases from all African countries have been reported. 4 The reasons 13 for the relatively lower numbers compared to other continents are unclear, but underreporting due to lack of diagnostic 14 capacity is likely a major factor limiting surveillance and patient care in Africa. 5 Therefore, the Chinese Jack Ma 15 Foundation donated more than 1 million RT-PCR-based SARS-CoV-2 test kits to all African countries in March 2020 16 (https://edition.cnn.com/2020/03/16/africa/jack-ma-donate-masks-coronavirus-africa/index.html; accessed 9 June 17 2020). Benin is a West-African country with 12 million inhabitants, representing one of the most densely populated African 19 regions ( Figure 1C ), which may facilitate intense SARS-CoV-2 spread. 6 The Laboratoire des Fièvres Hémorragiques 20 Virales du Benin (LFHB) is Benin's reference laboratory for respiratory diseases, testing all the country's suspected 21 COVID-19 cases. Here, we provide insight into all aspects of SARS-CoV-2 laboratory performance at LFHB, as an 22 example for the challenges inferred by the COVID-19 pandemic in Western Africa. Study design and participants 26 We assessed daily data on sample receipt and RT-PCR-based SARS-CoV-2 detections during January-April 2020. We characterized SARS-CoV-2 from 12 PCR-confirmed travelers entering Benin during March-April 2020 28 4 (appendix p 3). We obtained early convalescent sera from eight PCR-confirmed patients sampled during March-April 1 2020, taken an average 8 days post SARS-CoV-2 RT-PCR confirmation (range: 1-10 days) (appendix p 4). We 2 obtained 60 sera sampled from patients with acute febrile illness during hemorrhagic fever surveillance activities at 3 LFHB during October-November 2019 (appendix p 5-6; p 47). Samples were taken by trained technicians and stored 4 at -20°C until usage. This study was approved by the ethics committee of the Ministry of Health (Arrêté 2020 No. 5 030/MS/DC/SGM/DNSP/CJ/SA/027SGG2020) and followed the Declaration of Helsinki. Written consent was 6 obtained from all the patients participating in the study. Anonymized datasets were used, and all analysis of personally 7 identifiable data took place only in the LFHB. Time-stamped Bayesian phylogenies based on sampling dates were performed in BEAST2 (appendix p 2). 8 The 1 phylogeny was annotated with TreeAnnotator and visualized in FigTree from the BEAST package. Bayesian birth-2 death skyline analyses were performed in BEAST2 using parameters described in the appendix on p 2. Sera were tested using commercially available ELISAs relying on different antigens and antibody classes, namely 6 SARS-CoV-2 N antigen (IgG), spike S1 subunit (both IgG and IgA) and Middle East respiratory syndrome 7 coronavirus (MERS)-CoV S1 (IgG; all from Euroimmun, Germany). Additionally, sera were tested using 8 commercially available ELISA kits (Euroimmun) against the Zika virus (ZIKV) NS1 antigen (IgG), the Epstein-Barr 9 virus (EBV) EBNA1 antigen (IgG), the EBV VCA antigen (both IgM and IgG) and using real time-PCR for Plasmodia 10 (all human pathogenic species), EBV and cytomegalovirus (CMV) (all from TIB Molbiol). Plaque reduction 11 neutralization tests (PRNT) for SARS-CoV-2 and ZIKV were performed as described in the appendix on p 2. Antibody testing against common cold betacoronaviruses HCoV-OC43 and HCoV-HKU1 that belong to the same 13 viral genus as SARS-CoV-2, and may thus elicit cross-reactive antibodies, relied on recombinant spike protein-14 based immunofluorescence assays as previously described (appendix p 2). 9 15 16 Data availability The nucleotide sequences of the SARS-CoV-2 genomes used in this study are available at the GISAID database 18 (https://www.gisaid.org/) under accession IDs EPI_ISL_476822-EPI_ISL_476831 and EPI_ISL_476833-19 EPI_ISL_476834. One of the major problems that health systems worldwide face during the COVID-19 pandemic is the overload of 23 their diagnostic capacities. At the LFHB in Benin, oro-nasopharyngeal swabs for SARS-CoV-2 diagnostics were first 24 received the 1 st of March 2020 and the first case was detected in a traveler on the 14 th of March. Until the 28 th of April 25 2020, LFHB had received a total of 4,382 samples for SARS-CoV-2 molecular testing with up to 543 samples per day 26 (Figure 2) . The actual maximum testing capacity was up to 100 samples daily due to the limited availability of 27 personnel, reagents and laboratory equipment, which was exceeded on 14 days during March-April, demonstrating 28 6 the immense workload that LFHB has compensated using night shifts and all its available workforce only for SARS-1 CoV-2 diagnostics, at the cost of viral hemorrhagic fever surveillance. Notably, the average number of positive 2 samples detected per day was 1ꞏ4 (range 1-5), irrespective by the number of samples received per day, which may 3 have resulted from imprecise country-level case definitions. 10 4 5 By the 26 th of May 2020, 29,290 SARS-CoV-2 full genomes have been fully or partially sequenced by the global 7 scientific community and deposited in GISAID. Of those, only 1% (n=201) originated from Africa ( Figure 3A ). To 8 investigate the SARS-CoV-2 diversity introduced into Benin, we amplified 12 SARS-CoV-2 genomes from Beninese 9 citizens. One of these 12 individuals was a patient from a local hospital, whereas the other 11 were returning travelers 10 from Europe or Central-West African countries who showed no symptoms of disease but had to be tested for SARS- CoV-2 infection during quarantine upon re-entering Benin (appendix p 3). In a Bayesian phylogenetic analysis, the SARS-CoV-2 genomes clustered with the globally spreading SARS-CoV-2 13 lineages A and B 11 (Figure 3B) , which is in accordance with those individuals' travel history. To further characterize 14 the SARS-CoV-2 genomic diversity in Africa, we analyzed nucleotide differences in a dataset comprising the 15 Beninese and another 42 full African SARS-CoV-2 genomes available in GISAID until the 13 th of May. Within the 16 Benin-derived sequences, 12 variable nucleotide positions were observed, resulting in seven amino acid exchanges, 17 all of which were corresponding to previously published variable positions 12 ( Figure 3C ). According to distinct 18 genomic signatures, seven different clades were identified for SARS-CoV-2 in Africa (appendix p 7-9). The SARS- CoV-2 genomes from Benin belonged to three of those seven clades ( Figure 3C ). Those three clades also included 20 viruses from Algeria, the Democratic Republic of the Congo, Senegal and South Africa, as well as France and Italy, 21 hinting either at transmission across different African countries or parallel introduction of diverse SARS-CoV-2 22 lineages into African countries. In sum, genomic analyses suggest several independent introductions of globally 23 circulating SARS-CoV-2 lineages into Africa due to returning travelers, resulting in a high SARS-CoV-2 genetic 24 diversity in Africa. Bayesian skyline analyses to determine R0 27 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . Phylogenetic data can be used to estimate changes in the basic reproduction number (R0) over time. We inferred 1 R 0 using all SARS-CoV-2 genomes found in the GISAID databases from Africa until the 13 th of May and the novel 2 SARS-CoV-2 genomes from this study ( Figure 3D ). The calculated median R0 increased from 2.4 (95% confidence Diagnostics of acute SARS-CoV-2 infections rely on molecular testing. However, access to state-of-the-art reagents 10 and external quality control remain unresolved key issues of outbreak response in resource-limited regions. 13 In March 11 2020, Jack Ma, co-founder and head of the Alibaba Group in China, kindly donated and distributed more than 1 million 12 RT-PCR kits produced in China to Africa. No external validation of the kit has been performed to date, hindering 13 assessments of diagnostic performance. Using serial dilutions of quantified SARS-CoV-2 cell culture-derived RNA, 14 we determined a very high analytical sensitivity with a 95% limit of detection of 0ꞏ7-7ꞏ8 copies per reaction of both 15 assays included in this dual-target kit ( Figure 4A) . Notably, one of those two assays targeting the genomic ORF1ab 16 region showed a 10-fold lower sensitivity than the other assay targeting the N genomic region ( Figure 4A ). This may 17 lead to inconclusive results during testing of patient specimens containing low amounts of viral RNA sampled late 18 during the course of infection. 14 19 Nucleotide mutations in binding regions of PCR oligonucleotides are known to affect the diagnostic sensitivity of an 20 assay, potentially leading to false-negative results. 15 The exact target sites of the assays donated by the Jack Ma 21 foundation are unknown. Therefore, we tested the donated kit on six clinical swab samples representing the genetic However, in 29ꞏ6% (21/71) of the tested clinical sample replicates, the assay's internal control was not detected, which 25 according to the manufacturer's instructions invalidates the test. No further information about the internal control is 26 given in the manufacturer's protocol. In sum, the donated kit can be used confidently for diagnostics in Africa, but 27 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . 8 laboratories need to be alert about variations between the two different genomic targets and variations of the assay's 1 internal controls, which may have to be ignored to ensure diagnostic testing. 200,000 Chinese workers on the African continent (https://sais-cari.org/data-chinese-workers-in-africa). Serological 7 studies in European and Asian countries indicate high sensitivity and specificity of widely used SARS-CoV-2 8 serological tests such as ELISAs, but no assessment has been performed in African populations. 16, 17 In eight RT-PCR-9 confirmed patients from Benin, SARS-CoV-2 seroconversion ranged from 62ꞏ5 to 87ꞏ5% (7/8; 95% CI: 30ꞏ8-99ꞏ1%), 10 depending on the ELISA that was used and confirmed a higher sensitivity of the IgA-based compared to the IgG-11 based SARS-CoV-2 ELISAs ( Figure 5A ). 17 As shown in Figure 5B , 87ꞏ5% (7/8) of those ELISA results were 12 confirmed by a highly specific SARS-CoV-2 PRNT. In 60 samples taken during October-November 2019 from febrile 13 patients, 23ꞏ3% positive or borderline ELISA results potentially representing true positives 18 were observed (14/60; 14 95% CI: 14ꞏ3-35ꞏ5%). Different from PCR-confirmed cases, ELISA reactivity contrasted with the complete lack of 15 SARS-CoV-2-specific neutralizing antibodies in those samples ( Figure 5A and 5B). Likely unspecific SARS-CoV-2 16 ELISA reactivity may be consistent with three scenarios. First, antibodies elicited by common infections with endemic 17 human coronaviruses may cross-react with SARS-CoV-2 antigens. However, sera that yielded positive SARS-CoV-2 18 ELISA results did not differ significantly from sera that yielded negative SARS-CoV-2 ELISA results in their 19 reactivity with common cold coronaviruses (45ꞏ7-63ꞏ6% versus 70ꞏ4-74ꞏ0%; p=0ꞏ1 and p=0ꞏ7, Fisher´s exact test) 20 (appendix p 47). Similarly, the magnitude of antibody responses against common cold coronaviruses did not differ 21 significantly between those groups (p=0ꞏ09 and p=0ꞏ8, t-test) (appendix p 47). Second, polyclonal B-cell activation 22 can occur in infections or reactivations with herpesviruses such as CMV and EBV and elicit false-positive results in 23 serological tests. 19 However, only two patients were positive in a CMV PCR and one in an EBV PCR, and SARS-24 CoV-2 ELISA-positive versus ELISA-negative individuals did not differ in their past exposure to those human 25 herpesviruses according to detailed serological analyses (appendix p 48). Lastly, polyclonal B-cell activation can also 26 be caused by acute malaria, which is widespread in Africa 20 . As shown in Figure 5C , a higher proportion of those 27 individuals that yielded positive SARS-CoV-2 ELISA results than those that yielded negative ELISA results were 28 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . 9 positive for Plasmodia in a highly sensitive PCR test (71.4% versus 54ꞏ3%), albeit this difference was not statistically 1 significant (p=0.35, Fisher´s exact test). In contrast, significantly higher parasite loads occurred within SARS-CoV-2 2 ELISA-positive compared to ELISA-negative individuals ( Figure 5C ) (p=0.035; t-test). Higher parasite loads that 3 decrease overtime have been observed in acute malaria, suggesting a higher proportion of acute malaria in SARS-4 CoV-2 ELISA-positive patients compared to sub-acute or chronic malaria in SARS-CoV-2 ELISA-negative patients. 21 To assess the breadth of potentially unspecific reactivity, we tested the sera from febrile patients using a Zika virus 6 (ZIKV) IgG-ELISA for which unspecific reactivity in cases of acute malaria has been reported previously. 20 As shown 7 in Figure 5D , sera that elicited potentially unspecific SARS-CoV-2 ELISA results also elicited significantly more 8 frequently positive ZIKV ELISA results (57ꞏ1 versus 23ꞏ9%; p=0.019, Fisher´s exact test). None of the sera yielding 9 positive ZIKV ELISA results showed ZIKV-specific neutralizing antibodies, suggesting unspecific reactivity of those 10 sera in the ZIKV ELISA. Additionally, sera that yielded potentially false-positive results in the SARS-CoV-2 ELISA 11 were also significantly more likely to show potentially false-positive results in in the ZIKV ELISA (p=0.04; Chi-12 Square test) (appendix p 49). Notably, no serum reacted with MERS-CoV antigens, suggesting that unspecific 13 reactivity may not automatically apply to all coronavirus antigens and tests (appendix p 47). In sum, close to 25% of 14 the febrile patients showed unspecific reactivity in SARS-CoV-2 ELISAs, possibly due to acute malaria. We performed an observational study investigating COVID-19-related diagnostics in a West-African reference 18 laboratory. Our genome-based R 0 estimates, although preliminary, provide a blueprint to support notoriously weak 19 surveillance in resource-limited African settings and support increased efforts to characterize SARS-CoV-2 genomes 20 over time across geographic regions, similar to the large genomic dataset generated by multiple groups during the 21 2013-2016 West African Ebola outbreak. 22 Our relatively high R 0 estimate compared to other regions suggests intense 22 spread of SARS-CoV-2 on the African continent. 23 The relatively higher R0 compared to the initially low number of 23 reported cases from Africa likely results from weak surveillance systems, albeit a high number of asymptomatic 24 infections limiting accurate estimates of disease spread 24 or an insufficient genomic dataset to infer robust R0 estimates 25 cannot be ruled out. Nevertheless, Even an R 0 in the range of 2, which corresponds to the lower end of the 95% 26 confidence intervals of our R0 estimate, may imply more than 80 million cases in Africa if no intervention is 27 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . Overall SARS-CoV-2 daily diagnostic requests received at the LFHB since late January until 28.04.2020 (black) and 13 positive cases confirmed per day at the LFHB (red). Dotted lines denote the range of maximal daily diagnostic capacity 14 of LFHB. Marked is the 14.03.2020, day of the first confirmed SARS-CoV-2 case in Benin. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . Sangster All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . https://doi.org/10.1101/2020.06.29.20140749 doi: medRxiv preprint The basis for all generated datasets used in this study was a search for African SARS-CoV-2 full genome information on the Global Initiative on Sharing All Influenza Data (GISAID) platform on the 13th of May 2020. 1 To that date, 161 SARS-CoV-2 full genomes from African countries were available. Depending on the question behind the analysis, we generated three different datasets. 1. To analyze variable positions of all SARS-CoV-2 strains circulating in Africa, we generated a dataset discarding incomplete and identical sequences from the original GISAID search, resulting in 42 final sequences for analysis. 2. For the phylogenetic analysis we followed the nomenclature of Rambaut et. al. 2 and included one representative sequence of each of the 13 identified lineages A.1-A.5 and B.1-B.8 where fullgenome information was available in order to represent the global diversity of SARS-CoV-2 genomes. Finally, we completed this dataset by the full genomes we characterized from Benin. 3. For R 0 estimates we generated a third dataset expanding the Benin-derived genomes with all African sequences from the original GISAID search. Samples originated from three major health centers: Akkasato Health Center (AHC), Centre National Hospitalier Universitaire Hubert Koutoukou MAGA (CNHU) and the Clinique Boni (CB) inside or near of Cotonou (p 47). Plaque reduction neutralization tests were performed using similar methods for SARS-CoV-2 and for Zika virus. 3, 4 Briefly, VeroE6 cells were seeded in 24-well plates and incubated overnight. Prior to PRNT, patient sera were heat-inactivated at 56°C for 30 minutes. Next, patient sera were serially diluted in 200 µl OptiPro and mixed 1:1 with 200 µl virus solution containing 100 plaque forming units. The 400 µl serum-virus solution was vortexed and incubated at 37°C for 1 hour. Each 24-well was incubated 1 hour at 37°C with 200 µl serum-virus solution after which the cells were washed with PBS and supplemented with 1.2% Avicel solution in DMEM. After 3 days at 37°C, the supernatants were removed, and the 24-well plates were fixed and inactivated using a 6% formaldehyde/PBS solution and stained with crystal violet. Briefly, open reading frames for full spike proteins were cloned from human coronaviruses (HCoV) HCoV-OC43 and HCoV-HKU1 as described previously. 5 Vero cells were then seeded on chamber slides and transfected with equal amounts of the respective expression plasmids. Due to the limited quantity of serum available, samples were used at dilutions of 1:100, 1:1000, 1:10000. The following parameters and prior settings were chosen in BEAST2: SARS-CoV-2 was first detected late December of 2019, however, the earliest sequence from Africa in the GISAID database was sequenced in Nigeria on early February 2020 (hCoV-19/Nigeria/Lagos01/2020). As the origin date in BDSKY analysis has to be older than the root age, the lower bound was set to 114 days in the past from the last African sequence in the dataset, being end of April 2020 (Lognormal; M=0.8. S=1.0), to address a plausible earlier introduction to Africa in January 2020 assuming a month between introduction and detection, as suggested for SARS-CoV-2 first detection in China. 6 The "become uninfectious" rate (the reciprocal of the duration of infection) was set to 10 days using a lognormal distribution (M= -0.6, S= 1,25), based on the assumption that no viable virus was isolated after 10 days post onset of symptoms. 5 Priors for the sampling proportion followed the consideration that our alignment consists only of a tiny fraction of the circulating SARS-CoV-2 in Africa (alpha=1, beta=9999). All analyses were performed with a relaxed clock exponential using recent SARS-CoV-2 evolutionary rate estimates of 1E-4 substitutions per site per year. 6 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020 . . https://doi.org/10.1101 Supplementary All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020 . . https://doi.org/10.1101 Supplementary (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . EPI_ISL_437337|2020-04-15 EPI_ISL_437348|2020-04-16 EPI_ISL_435032|2020-04-09 EPI_ISL_417437|2020-03-17 EPI_ISL_420030|2020-03-21 EPI_ISL_420847|2020-03-26 EPI_ISL_417947|2020-03-19 EPI_ISL_437350|2020-04-16 EPI_ISL_420032|2020-03-22 DR Congo All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . https://doi.org/10.1101/2020.06.29.20140749 doi: medRxiv preprint 8 EPI_ISL_437352|2020-04-17 EPI_ISL_437339|2020-04-15 EPI_ISL_420070|2020-03-17 South Africa 03-02 EPI_ISL_418242|2020-03-08 260/Benin|2020-03-17 Benin All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . https://doi.org/10.1101/2020.06.29.20140749 doi: medRxiv preprint EPI_ISL_420033|2020-03-22 EPI_ISL_428855|2020-03-17 *Nucleotide positions that are also variable within Benin sequences. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . https://doi.org/10.1101/2020.06.29.20140749 doi: medRxiv preprint We gratefully acknowledge the following Authors from the Originating laboratories responsible for obtaining the specimens and the Submitting laboratories where genetic sequence data were generated and shared via the GISAID Initiative, on which this research is based. All submitters of data may be contacted directly via www.gisaid.org (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. C) Relation of reactivity between SARS-CoV-2 S1-IgA, S1-IgG and N-IgG ELISA positive patients with ZIKV-IgG ELISA. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted July 8, 2020. . William Ampofo EPI_ISL_428855 hCoV-19/Gambia/GC19-015/2020 2020-03-17 MRCG at LSHTM Geomics lab MRCG at LSHTM Genomics lab Sesay et al EPI_ISL_428856 hCoV-19/Gambia Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Steve Ahuka-Mundeke, Jean-Jacques Muyembe Tamfum All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Global initiative on sharing all influenza data -from vision to reality A dynamic nomenclature proposal for SARS-CoV-2 to assist genomic epidemiology High Zika Virus Seroprevalence in Salvador, Northeastern Brazil Limits the Potential for Further Outbreaks Severe Acute Respiratory Syndrome Coronavirus 2-Specific Antibody Responses in Coronavirus Disease Virological assessment of hospitalized patients with COVID-2019 No evidence for increased transmissibility from recurrent mutations in SARS-CoV-2 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted July 8, 2020.