key: cord-0255713-lboi0d4a authors: Zhao, B.; Yu, Z.; Fujita, T.; Nihei, Y.; Tanaka, H.; Ihara, M. title: Tracking community infection dynamics of COVID-19 by monitoring SARS-CoV-2 RNA in wastewater, counting positive reactions by qPCR date: 2021-12-24 journal: nan DOI: 10.1101/2021.12.23.21268343 sha: 21fe4ee0c5d60130e63d73adc2075693073e55b0 doc_id: 255713 cord_uid: lboi0d4a Wastewater-based epidemiology has proved useful for monitoring the COVID-19 infection dynamics in communities. However, in some countries, low concentrations of SARS-CoV-2 RNA in wastewater make this difficult. Getting meaningful information from wastewater-based epidemiology in regions of low prevalence remains a key challenge. Here we used real-time reverse-transcription PCR (RT-qPCR) to monitor SARS-CoV-2 RNA in wastewater from October 2020 to February 2021 during the third wave of the COVID-19 outbreak in Japan. Viral RNA was below the limit of quantification in all samples. However, by counting the positive reactions in repeated qPCR of each sample, we found that the ratio of positive reactions to all tests in wastewater was significantly correlated with the number of clinically confirmed cases by the date of symptom onset during periods of both increasing and decreasing infection. Time-step analysis indicated that COVID-19 patients excreted large amounts of virus in their feces 2 days either side of symptom onset, which wastewater surveillance could detect. The positive count method is thus useful for tracing COVID-19 dynamics in regions of low prevalence. analysis. However, in many countries, the spread of the virus has exceeded testing capacity, 44 precluding real-time monitoring of the pandemic. Instead, wastewater-based epidemiology 45 (WBE) has been considered as an effective approach for monitoring the presence of CoV-2 in the community. 2 47 Increases and decreases of viral RNA concentrations measured by real-time reverse-48 transcription PCR (RT-qPCR) in raw influent or its solid fraction have been associated with 49 those of COVID-19 cases in Australia 3 , the Netherlands 4 , the USA 5-8 , and Canada 9,10 . In 50 Japan, some studies have detected SARS-CoV-2 RNA in wastewater when cases were 51 diagnosed by clinical testing in the community. [11] [12] [13] [14] Concentrations in the solid fraction of raw 52 influent were quantifiable, 11 but the supernatant fraction had high PCR threshold cycle (Ct) 53 values, and most data were unquantifiable owing to low concentrations of SARS-CoV-2 RNA 54 in the wastewater. 12-14 Getting meaningful information from WBE in regions of low 55 prevalence remains a key challenge. 56 In general, the limit of quantification (LOQ) of qPCR for the virus in wastewater is 57 based on the fewest copies of the control molecule in a reaction volume that can be 58 quantitatively determined with a stated probability; e.g., 5 or 10 copies per reaction. For 59 wastewater with Ct smaller than those of 5 or 10 copies, the virus RNA copy number is 60 titrated. These wastewater is rated either positive or negative for the virus RNA relative to the 61 All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 LOQ. In qPCR, when the number of initial target molecules is <10, the probability that an 62 aliquot contains a given number of target molecules is given by a Poisson distribution, with 63 which the pattern of positive and negative results is consistent. [15] [16] [17] In this range, the chance of 64 a positive increases with the number of initial target molecules in the solution. Thus, we 65 hypothesized that by calculating the positive ratio (the ratio of positive reactions to all tests) 66 in repeated qPCR analysis of wastewater, we might track the infection dynamics even in 67 regions of low prevalence. 68 The purpose of this study was to verify the effectiveness of calculating the positive ratio 69 for SARS-CoV-2 RNA in wastewater to track the infection dynamics of COVID-19 in the 70 community. First, we tested the pattern of positive and negative results in qPCR in the range 71 of 0.1 to 20 initial copies of oligo DNA including the CDC-N1 target per reaction, and 72 confirmed that empirical data obtained by qPCR of these samples followed a Poisson 73 distribution. Then we took raw or primary effluent from wastewater treatment plant (WWTP) 74 A in city B in the Kansai region of Japan almost every day from October 2020 to February 75 2021, during the third wave of COVID-19 in Japan. After polyethylene glycol (PEG) 76 precipitation to concentrate the virus, we analyzed SARS-CoV-2 RNA in the wastewater by 77 RT-qPCR. The positive ratio for SARS-CoV-2 RNA in technically repeated qPCR was 78 compared with epidemiological data in the community reported by the public health system. 79 In most samples, SARS-CoV-2 RNA was below the LOQ, but the positive ratio in repeated 80 qPCR for N1 and N2 assays in wastewater was significantly correlated with the number of 81 daily confirmed cases by the date of symptom onset. Time-step analysis indicates that 82 COVID-19 patients excreted a large amount of virus in their feces during the period from 2 83 days before to 2 days after symptom onset, which wastewater surveillance could detect. 84 The Poisson distribution for PCR is defined as: where e is the base of the natural logarithm, C is the average initial target molecule number 90 (ITMN) per qPCR sample (expected ITMN), and k is the actual number (actual ITMN) in a 91 sample. P is the probability that a volume (sample) contains k copies of ITMN. SI Figure S1 92 All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 shows the probabilities of obtaining a certain number of target molecules in a given volume, 93 where the sample average concentration ranges from 10 to 0.01 copies per volume. 94 qPCR assays for low target copy number were performed. Oligo DNA including the 95 SARS-CoV-2 sequence (Thermo Scientific) was diluted to prepare different ITMN samples 96 (C = 0.01, 0.1, 0.2, 0.4, 0.6, 0.8, 1, 1.2, 1.4, 1.6, 1.8, 2, 4, 6, 8, or 10 copies per reaction), and 97 the CDC 2019-nCoV_N1 assay 18 was performed 12 times for each ITMN sample. Details of 98 the qPCR for SARS-CoV-2 are described below under "qPCR for SARS-CoV-2". We 99 counted the positives in the 12 qPCR reactions for each ITMN sample, and then compared the 100 resulting "positive ratios" with those expected from the Poison distribution. These 101 experiments were performed three times (three batches). 102 Influent or primary effluent (PE) was collected at WWTP A in city B. PE was collected two 104 times (Tuesday and Friday) a week from 20 October to 14 December 2020, and five times 105 (Monday to Friday) a week from 15 December 2020 to 14 January 2021. Influent was 106 collected five times (Monday to Friday) a week from 15 January to 15 February. In total, 60 107 samples (39 PE and 21 influent) were collected in the morning (09:00-10:00). 108 In October 2020, clinical testing in city B confirmed few to no daily COVID-19 cases. 109 From November 2020 to February 2021, city B experienced an increase followed by a 110 decrease in the number of cases, during the third wave of infection in Japan. 111 Wastewater was collected in a sterilized 250-mL plastic bottle, immediately transported 112 to the laboratory, and then heat-treated in water bath (60 °C, 90 min) to inactivate the 113 coronavirus. 19, 20 The heat-treated samples were stored at −30 °C. Within a week after 114 collection, they were analyzed for SARS-CoV-2 RNA by RT-qPCR. 115 Virus in each sample was concentrated by PEG precipitation 13 with a slight modification. 117 First, 120 mL of PE was centrifuged at 4500× g for 10 min, and the supernatant was 118 transferred to a fresh centrifuge tube. Then, PEG 8000 (Molecular Biology Grade, average 119 mol wt 8000; Sigma-Aldrich) and NaCl were added to final concentrations of 10% and 1 M, 120 respectively. The samples were incubated at 4 °C overnight with gentle agitation. After 121 centrifugation at 10 000× g for 30 min, the PEG precipitate, containing the virus, was 122 dissolved in 500 μ L of phosphate buffer solution (for biochemistry, 0.1 M, pH 8.0, Wako) to 123 give a total volume of ~700 μ L. From 140 μ L of the virus concentrate, RNA was extracted 124 All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. The SARS-CoV-2 RNA was assayed by the CDC 2019-nCoV_N1 and CDC 2019-nCoV_N2 134 qPCR assays 18 in all samples with the primers and probes listed in Table S1 . qPCR assays in 135 96-well plates were conducted in a 25-μL qPCR reaction volume, which included 12.5 Kusatsu, Shiga, Japan). The thermal cycling conditions for both assays consisted of pre-141 heating at 50 °C for 2 min and pre-denaturation at 95 °C for 10 min, followed by 50 cycles of 142 amplification at 95 °C for 15 s, and annealing and extension at 60 °C for 1 min. For each 143 sample, qPCR was performed in technical triplicate for each primer set, and each test was 144 performed twice. Therefore, 6 reactions were obtained for each N1 and N2 assay of each 145 sample. 146 Every qPCR assay included a negative control (PCR-grade water), and no amplification 147 was observed in either assay. There was no amplification in the negative controls for either 148 the RNA extraction step or the RT step. 149 As a positive control, a 10-fold serial dilution of double-stranded oligo DNA including 150 both CDC 2019-nCoV_N1 and CDC 2019-nCoV_N2 targets (GeneArt Strings DNA 151 Fragments; Thermo Scientific) was used to generate standard curves (from 10 1 to 10 4 copies 152 per 5 μ L). The LOQ was 10 copies per reaction with Ct values of 37.7 ± 1.83 for N1 and 38.6 153 ± 2.28 for N2 primer sets. The theoretical LOQ of the overall method was 4.0 log 10 copies/L 154 for N1 and N2. The N1 primer set generated a standard curve with R 2 = 0.98 ± 0.04 (n = 27 155 reactions) with an efficiency (mean ± SD) of 106% ± 13.4% (slope = −3.22 ± 0.274; y 156 intercept = 40.9% ± 2.01%). The N2 primer set generated a standard curve with R 2 = 0.97 ± 157 All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101/2021.12.23.21268343 doi: medRxiv preprint 6 0.05 (n = 27 reactions) with an efficiency of 97.7% ± 14.7% (slope = −3.43 ± 0.373; y 158 intercept = 42.0% ± 2.54%). 159 SARS-CoV-2 RNA was detected in many samples in both N1 and N2 qPCR assays, always 161 below the LOQ (see Results and Discussion). Therefore, instead of quantification for SARS-162 CoV-2 RNA genome copy number, we counted the number of positives in 6 repeated qPCR 163 reactions for each assay. Maximum Ct values were 45 or 46 (Results and Discussion). 164 PMMoV is the most abundant human fecal RNA virus, 21 and has recently been used as an 166 internal control for SARS-CoV-2 in wastewater. 7, [9] [10] [11] 20 When fluorescence reached the 167 threshold during 50 cycles of qPCR in each well of the 96-well plate, we counted it as a 168 positive. We also tested PMMoV RNA by RT-qPCR in the same samples, using 5 μ L of 169 template cDNA for all samples, in duplicate, with the primers and probes shown in Table S1 170 and the same thermal cycling conditions as for SARS-CoV-2 qPCR. We monitored Ct values 171 in each sample to check any loss of SARS-CoV-2 RNA detection. 172 Agarose gel electrophoresis of some samples confirmed the qPCR amplicon sizes: 72 bp for 174 CDC 2019-nCoV_N1 and 67 bp for N2. In addition, qPCR products of N1 and N2 assays of 175 samples 25 and 33 were sequenced by the Molecular Biology Laboratory (Unitech Co., Ltd.; 176 Kashiwa, Chiba, Japan) to confirm whether the qPCR amplified target sites of SARS-CoV-2 177 RNA. The genome sequence of SARS-CoV-2 Wuhan-Hu-1 strain (GenBank accession No. 178 MN908947.3) was used as a reference. 179 All patients confirmed positive for COVID-19 by clinical testing in city B were 181 retrospectively interviewed by local public health officers to complete contact tracing. 182 Officers collected recorded symptoms, symptom onset date, contact with other known 183 clinically confirmed cases, and the date of reporting clinical test result (i.e., the date 184 confirmed positive). These data are publicly available. We counted the daily number of cases 185 by the date of symptom onset and by the date reported, from October 2020 to February 2021 186 (see Results and discussion). For positive asymptomatic cases, the date on which the test 187 result was reported was used as the estimated date of symptom onset. 188 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We investigated whether the SARS-CoV-2 RNA signals in wastewater samples correlated 190 with the epidemiological data. For each wastewater sampling day from October 2020 to 191 February 2021, we compared the number of positives in SARS-CoV-2 RNA PCR reactions 192 and the number of cases who developed symptoms on the same day. We also compared the 193 number of positives in wastewater with the number of cases in clinical testing reported on the 194 same day. Spearman's rank correlation test was performed in GraphPad Prism 5 software. All 195 statistical tests were two-tailed, with P < 0.05 considered as statistically significant. 196 To evaluate any lag between the SARS-CoV-2 RNA signal in wastewater and the 197 epidemiological data, we performed time-step analyses; we compared the number of positives 198 in wastewater with the number of cases who developed symptoms from 10 days before to 10 199 days after the wastewater sampling day (i.e., time lag = −10 to +10). 200 distribution 203 qPCR assays of low-copy-number samples revealed positive reactions in the 12 technical 204 replicates of each ITMN sample (n = 12), and positive ratios for each ITMN sample were 205 calculated ( Figure 1) . Ct values gradually increased as the initial copy number of oligo DNA 206 of SARS-CoV-2 decreased (Figure 1, left) . At the same time, the positive ratio decreased. All 207 reactions were positive (n = 12, 100%) in all 3 batches when ITMN was ≥ 6 copies per 208 reaction; around 90% were positive when ITMN was 2 copies; and around 0% to 17% when it 209 was 0.2 copies (Table S2 ). These positive ratios for each ITMN agree well with those 210 expected from a Poison distribution (Table S3 ; Figure 1 , right). Similar results were reported 211 in previous studies. 15 From these results, we conclude that by counting the number of positive 212 reactions in repeated qPCR assay, we could trace the dynamics of SARS-CoV-2 RNA in 213 wastewater in the range of 0.1 to 4 copies per reaction of ITMN, which is usually below the 214 LOQ in qPCR and is thus regarded as negative. 215 PMMoV was tested as an internal control in all 60 wastewater samples. It was detectable and 225 stable between daily samples, with Ct ranging from 25 to 29 (Figure 2, top; SI Table S4 ). This 226 result indicates that sample collection, PEG precipitation, RNA extraction, and qPCR 227 processes were not exceptional throughout the sampling campaign. 228 Maximum Ct values were around 45 in the N1 assay and 46 in the N2 assay (Figure 2 , 229 bottom; Table S5 ). There was no positive detection on the first 3 sample collection days (20, 230 23, and 27 October). The first positive was detected on 30 October, when the N1 assay (n = 6) 231 gave 1 positive and the N2 assay (n = 6) gave all negatives. 232 During early to mid November, both assays gave several positives in repeated PCR 233 reactions on all sampling days, with Ct values of 40 to 45 in the N1 assay and 42 to 45 in the 234 N2 assay (Figure 2, bottom) . In late November, the number of positives decreased. But from 235 December, it then increased until mid January 2021, and then again decreased until 15 236 February. During this period, Ct values were 38 to 45 in the N1 assay and 39 to 46 in the N2 237 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 244 the sample showed at least 1 negative (n = 6). Ct values are shown in SI Table S5 . Gel electrophoresis confirmed the sizes of the PCR products to be identical to the target 246 sizes (72 bp by N1, 67 bp by N2) by (data not shown). Partial nucleotide sequences of qPCR 247 products from samples 25 and 33 (Table S4) (Table S5 ; Figure 3 , bottom panel, N1 + N2). It showed a 257 small peak in November 2020 and a bigger and more persistent peak from December to 258 January (Figure 3) , when it sometimes reached the maximum number of positives, i.e., 12. In 259 February, it decreased to as low as 0. 260 All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 Comparison of viral signals in wastewater with the number of new cases (Table S6) All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101/2021.12.23.21268343 doi: medRxiv preprint Table S6 . To evaluate any lag between the SARS-CoV-2 RNA signal in wastewater and the 285 epidemiological data, we performed time-step analyses, comparing the number of positives in 286 wastewater in repeated qPCR reactions on each sampling day with the number of new cases 287 by the date of symptom onset from 10 days before to 10 days after the wastewater sampling 288 day ( Figure S4 ). The number of new cases 2 days before ( Figure 5 , time lag = −2 days, 289 Spearman's r = 0.7753) and 1 day before (−1, r = 0.7279) the sampling day had higher 290 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Nov.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Dec.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Jan.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 1 2 correlations than earlier days (from −10 to −3). And the number of new cases 1 day after (+1, 291 r = 0.7177) and 2 days after (+2: r = 0.7283) the sampling day had higher correlations than 292 later days (from +3 to +10, r < 0.7). These results indicate that COVID-19 patients excreted 293 more virus in their feces in the period from 2 days before to 2 days after symptom onset than 294 on other days, which wastewater surveillance could detect. 295 Previous studies have reported that SARS-CoV-2 RNA concentrations in wastewater 296 were correlated with epidemiological data such as the number of clinically confirmed cases 297 by the date of specimen collection or of test result reporting, 4-8,10 but did not investigate the 298 correlation with the number of cases by the date of symptom onset. So far, only one study has 299 investigated this correlation, and reported that wastewater detection of the SARS-CoV-2 RNA All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. How much sooner did the SARS-CoV-2 RNA concentration in wastewater than the date of 315 result reporting or specimen collection in the public health system? On average, a person 316 develops symptoms 4 to 5 days after infection. 22 Our results indicate that SARS-CoV-2 RNA 317 concentrations in feces are highest during 2 days either side of symptom onset ( Figure 6 ). In 318 regions where the time lag between the date of symptom onset and the date of clinical test 319 result reporting is large, WBE, if analyzed and reported on the same day as sampling, will 320 give the earliest estimate of viral spread in the community. This is why WBE has an 321 advantage over clinical testing in the public health system. 322 323 Figure 6 . Timeline of clinical testing and wastewater surveillance. Throughout the sampling period, N1 and N2 assays in wastewater samples showed significant 326 correlation with the number of new cases in the range of 0 to 30 (Figure 4 ). The population of 327 city B is about 300,000, and WWTP A serves about one-third of it. This means that SARS-328 CoV-2 RNA in wastewater could track the infection dynamics in the community even when 329 the number of new cases ranges from 0 to 10 per 100,000 population. However, the 330 possibility of asymptomatic patients means that the actual number of cases might be larger. 331 These results indicate that COVID-19 patients excrete more virus in their feces during 2 days 333 either side of symptom onset than earlier or later ( Figure 5 ). Clinical studies have reported 334 virus concentrations in feces several days after symptom onset 23 or after admission to 335 hospital 24, 25 . Under the assumption that virus shedding in feces starts from the time of 336 symptom onset, re-analysis of patient data using a shedding dynamics model indicated that 337 the SARS-CoV-2 RNA concentration in feces increases rapidly after symptom onset. 26 338 However, no clinical data on viral shedding in feces from infection to symptom onset are 339 reported. Still, it is possible that the peak occurs before symptom onset. To understand what 340 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. the viral RNA signal in wastewater represents, the peak timing of viral shedding in feces is 341 essential information. More fecal shedding data are needed. 342 As patients likely shed SARS-CoV-2 in feces for weeks, 24, 26, 27 it is somewhat surprising 343 that SARS-CoV-2 RNA signals in wastewater significantly correlated with daily confirmed 344 new cases in this study. But these results indicate that the magnitude of fecal shedding might 345 be substantially higher around the day of symptom onset. Another possibility is that the virus 346 RNA in wastewater is derived not only from feces, but also from the respiratory system and 347 saliva: virus excreted during brushing the teeth, gargling, and rinsing could mix with virus in 348 feces and travel to the WWTP. Viral concentrations in throat swabs from patients were 349 highest immediately after onset, and therefore it is estimated that infectiousness is highest 350 from 2 to 1 day before the onset date. 28 The period of highest infectiousness might also be the 351 period of highly detectable virus RNA in wastewater. 352 Although SARS-CoV-2 RNA concentrations in wastewater were lower than the LOQ, the 354 number of positives in repeated RT-qPCR reactions could trace the prevalence of in the community. In conventional qPCR testing, the limit of quantification is usually 5 to 10 356 copies per reaction. In contrast, the positive count method could distinguish lower copy 357 numbers, i.e., 0.1 to 4 copies per reaction, which means that it increases the sensitivity of 358 qPCR by 50 to 100 times to reveal the trend of viral RNA concentrations in wastewater. concentrating virus from the solid fraction of raw influent or primary sludge rather than the 366 water fraction. 6, [8] [9] [10] [11] However, in regions where the virus concentration in wastewater is not 367 high enough for quantification from the solid fraction, the positive count methods might be 368 useful to reveal changes in viral RNA levels in wastewater. 369 The idea of the positive count method is the same as that for the quantification of droplet 370 digital PCR. Therefore, we expect to see use of the droplet digital PCR in future studies to 371 show the trend of SARS-CoV-2 RNA at low copy numbers in wastewater, as the positive 372 count method has done here. 373 Sewage-For-Distribution SARS-CoV-2 titers in wastewater are higher than 456 expected from clinically confirmed cases Pepper mild mottle virus as a water quality 458 indicator The incubation period of coronavirus disease from publicly reported confirmed cases: estimation and application Virological assessment of 466 hospitalized cases of coronavirus disease 2019 SARS-CoV-2 infection and potential evidence for persistent fecal viral shedding All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. CoV-2 in wastewater in Japan during a COVID -19 outbreak. Sci. Total Environ. 2021, 758, 436 143578. 437 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. -pcr-panel-primer-probes.html. 2020 -pcr-panel-primer-probes.html. . (accessed December 17, 2021 . 451All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101 /2021 doi: medRxiv preprint shedding and transmissibility of COVID-19. Nat. Med. 2020, 26, 672-675. 486 All rights reserved. No reuse allowed without permission. preprint (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 this version posted December 24, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021