key: cord-318387-s4d442kx
authors: Wang, Ming; Fu, Aisi; Hu, Ben; Tong, Yongqing; Liu, Ran; Gu, Jiashuang; Liu, Jianghao; Jiang, Wen; Shen, Gaigai; Zhao, Wanxu; Men, Dong; Yu, Lilei; Deng, Zixin; Li, Yan; Liu, Tiangang
title: Nanopore target sequencing for accurate and comprehensive detection of SARS-CoV-2 and other respiratory viruses
date: 2020-03-06
journal: nan
DOI: 10.1101/2020.03.04.20029538
sha: 
doc_id: 318387
cord_uid: s4d442kx

The ongoing novel coronavirus pneumonia COVID-19 outbreak in Wuhan, China, has engendered numerous cases of infection and death. COVID-19 diagnosis relies upon nucleic acid detection; however, current recommended methods exhibit high false-negative rates, low sensitivity, and cannot identify other respiratory virus infections, thereby resulting patient misdiagnosis and impeding epidemic containment. Combining the advantages of target amplification and long-read, real-time nanopore sequencing, we developed nanopore target sequencing (NTS) to detect SARS-CoV-2 and other respiratory viruses simultaneously within 6-10 h. Parallel testing with approved qPCR kits of SARS-CoV-2 and NTS using 61 nucleic acid samples from suspected COVID-19 cases confirmed that NTS identified more infected patients as positive, and could also monitor for mutated nucleic acid sequence or other respiratory virus infection in the test sample. NTS is thus suitable for contemporary COVID-19 diagnosis; moreover, this platform can be further extended for diagnosing other viruses or pathogens.

An ongoing novel coronavirus pneumonia 

December 2019 has subsequently spread across China and worldwide, resulting in numerous cases 40 of infection and death 1 . Usually, COVID-19 has an incubation period of 2-7 days 2 with no obvious 41 symptoms, during which time the virus can spread from infected to uninfected individuals. 42 Therefore, early accurate diagnosis and isolation of patients is key to controlling the epidemic. 43

Nucleic acid detection is the golden standard for COVID-19 diagnosis. Real-time reverse 44 transcription-polymerase chain reaction (qPCR) is the most recommend testing method for 45 detecting the causative virus, SARS-CoV-2 3 . qPCR is specific, rapid, and economic, but cannot 46 precisely analyze amplified gene fragment nucleic acid sequences; thus, positive infection is 47 confirmed by monitoring one or two sites (depending on manufacturer guidelines). However, qPCR 48 exhibits high false-negative rates and low sensitivity in clinical application 4 , with only 30-50% 49 positive detection ratio. False-negatives facilitate epidemic spread through delayed patient isolation 50 and treatment, and patients mistakenly considered uninfected or cured following misdiagnosed 51 treatment results. Another recommend detection method, sequencing, is widely applied for 52 pathogen identification and monitoring virus evolution 5, 6 including SARS-CoV-2 7 , but requires 53 expensive equipment, operator expertise, and > 24 h turnaround time, rendering it unsuitable for the 54 current crisis. 55

Several intelligent methods for RNA virus detection have developed including combining 56 toehold switch sensors 8 , which can bind to and sense virtually any RNA sequence, with paper-based 57 cell-free protein synthesis for Ebola and Zika virus detection 9, 10 , and the SHERLOCK method 58 based on CRISPR/Cas13a for Zika or Dengue virus detection 11 . A rapid SHERLOCK method with 59 visual results can detect SARS-CoV-2 12 and toehold switch biosensors could theoretically be 60 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is AA), and ORF10 (38 AA) proteins. We also considered the RNA-dependent RNA polymerase 86 (RdRP) region in orf1ab (Fig. 1) . For the virulence regions, 11 fragments of 600-950 bp were 87 designed as targets, fully covering the 9,115 bp region (Fig.1) , amplified by 22 specific primers 88 designed considering primer-primer interaction and annealing temperature, and potential non-89 specific binding to human and common bacterium and fungi genomes. To improve the sensitivity 90 orf1ab region amplification, we designed two pairs of nested primers to amplify 300-500 bp 91 regions to avoid amplification failures owing to site mutation. Finally, the 26 primers were 92 combined to develop the SARS-CoV-2 primer panel (Supplementary Table 1) . 93

For sequencing, we chose a nanopore platform that could sequence long nucleic acid fragments 94 and simultaneously analyze the data-output in real-time. This allowed confirmation of SARS-CoV-95 2 infection within a few minutes after sequencing by mapping the sequence reads to the SARS-96

CoV-2 genome and analysis of output sequence identity, coverage, and read number. Moreover, the 97 accurate nucleic acid sequence generated using our pipeline could indicate whether the virulence-98 related genes were mutated during virus spreading, thereby rapidly providing information for 99 subsequent epidemiological analysis. Additionally, as the MinION nanopore sequencer is portable, 100 NTS is also suitable for front-line clinics. 101 102 NTS results interpretation and limit of detection (LoD). To test the SARS-CoV-2 detection 103 efficiency by NTS, we used standard plasmids harboring COVID-19 virus S and N genes to 104 simulate SARS-CoV-2. Standard plasmids were individually spiked into background cDNA 105 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 6 samples (cDNA reverse-transcribed from an uninfected respiratory flora throat swab) at 10, 100, 106 500, 1000, and 3000 copies/mL. NTS for all test samples was performed on one MinION sequencer 107 chip. Sequence data were evaluated at regular intervals using our in-house bioinformatics pipeline. 108

By mapping output reads on the SARS-CoV-2 genome, all reads with high identity were calculated 109 for each plasmid concentration. For 10 min and 1 h sequencing data, reads mapped to SARS-CoV-2 110 significantly differed from those of negative controls in all replicates at concentrations ranging from 111 3000 to 500 (Fig. 2a) , and 3000 and 10 ( Fig. 2b ) copies/mL, respectively. This result confirmed that 112 high-copy samples could rapidly yield sufficient valid sequencing data for diagnosis, and by 113 extending the sequencing time, valid sequencing data could also be obtained from low-copy 114 samples. Notably, as more sequencing data could be achieved with additional sequencing time 115 Evaluation of the target distribution of these valid data revealed that in higher copies samples 119 (1000 and 3000 copies/mL), all targeted regions could be detected (Fig. 2c, d) . However, in lower 120 viral concentration samples (from 10 to 500 copies/mL), some targeted regions were lost (i.e., no 121 reads mapped; Fig. 2c , d), indicating that for low-quality or low-abundance samples, comprehensive 122 fragment amplification is difficult. Therefore, for accurate results, NTS cannot label a sample as 123 positive for infection by monitoring only one or two sites, as is customary for qPCR; rather, the 124 results from all target regions should be considered. 125

Accordingly, we determined a scoring rule by referring to previous judgment rules 14-16 . Firstly, 126

we counted the number of output reads with high identity to the SARS-CoV-2 genome, indicative 127 of high credibility for identification as SARS-CoV-2. By calculating the ratio of the counted valid 128 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint read numbers of the test sample to those of the negative control (with "0" in the negative control 129 calculated as "1"), we defined that a ratio of ≥10 indicates a positive result for that fragment, 130 scoring 1; ≥3 to 10 fold is inconclusive, scoring 0.4; and <3 is negative, scoring 0. Scores were 131 summed to obtain the NTS score. We considered that a sample in which at least 50% fragments (6 132 fragments) are inconclusive or 2 fragments are positive (comparable to qPCR results) could be 133 defined as a positive infected sample (e.g., NTS score >2.4); 3-6 inconclusive fragments or 1 134 positive fragment indicated a highly suspect (inconclusive) sample (e.g., NTS score of 1.2-2.4); and 135 < 3 inconclusive or no positive fragments could be defined as negative sample (NTS score <1.2). 136

To determine the NTS LoD, we used the defined rules to evaluate each replicate in the 137 simulated test. As the standard plasmids contain only 6 designed fragments (half of 12 designed 138 fragments for SARS-CoV-2), we defined the scoring as NTS score >1.2 indicates positive 139 detection, 0.6-1.2 is inconclusive, and < 0.6 reflects negative detection. We calculated the score of 140 the lowest concentration (10 copies) at different times according to this scoring method and judged 141 the positive detection rate. The results (Supplementary table 2) showed that 3/4 of the 10 copies of 142 the standard plasmids can be judged positive from 1 h. This result is consistent with the significant 143 comparation (Fig. 2b) , that the data for 10 copies standard plasmids is also significantly different 144 from the negative control from 1h. This result shows that our scoring system is reliable for 145 evaluating NTS test results, and the LoD (3/4 replicates positive) was determined as 10 copies/mL 146 with 1h sequencing data (1,372 to 43,967 reads per sample in a run with 24 samples). 147 148 SARS-CoV-2 detection using qPCR vs NTS. We performed clinical sample testing at the first-149 line hospital in Wuhan as soon as NTS method was established (Fig. 3) . To verify NTS sensitivity, 150

we evaluated 45 nasopharyngeal swab samples from outpatients with suspected COVID-19 early in 151 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 8 the epidemic. On February 6 and 7, 2020, we parallel tested these 45 samples in two batches using 152 NTS (two chips) and qPCR (kit 2; Fig. 1 ). The 4 h sequencing output data (Fig. 4a ) revealed that all 153 19 samples defined as positive by qPCR were recognized SARS-CoV-2-infected by NTS, 154

indicating good inter-test concordance. Among 15 qPCR-inconclusive samples, 11 were recognized 155 as SARS-CoV-2-infected, 3 as negative, and 1 inconclusive by NTS. Among 11 qPCR-negative 156 samples, 4 were recognized SARS-CoV-2-infected, 4 as inconclusive, and 3 as negative by NTS. qPCR. Three positive samples were identified by 10 min sequencing ( Supplementary Fig. 2) , 172

indicating that NTS could rapidly detect positive samples with high concentration of virus. 173 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint Evaluation of the positive target distribution for each sample (Fig. 4) the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 11 community and used to test the NTS virus detection capacity. NTS 10 min sequencing data 220 (Supplementary Table 5 Herein, we developed an NTS method able to simultaneously detect SARS-CoV-2 and 10 237 additional types of respiratory viruses within 6-10 h, at LoD of 10 copies/mL with at least 1 h 238 sequencing data. The detection region of SARS-CoV-2 was composed of 12 fragments covering 239 nearly 10 kb of the genome, resulting in markedly higher sensitivity and accuracy than those of 240 qPCR kits currently in clinical use. Notably, 22 of 61 suspected COVID-19 samples that were 241 negative or inconclusive by qPCR testing were identified as positive by NTS. Moreover, NTS 242 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint enabled the detection of virus mutations; in particular, we detected a nucleotide mutation in SARS-243

CoV-2 that was undetected in the genomic data in the current GISAID database. Although this was 244 a silent mutation, its presence suggests that the virus may have undergone mutation during the 245 spreading process. Additionally, NTS was verified as capable of detecting all five pre-added 246 respiratory viruses in a single detection. This method also detected a co-infected case (SARS-CoV-247 2 and human influenza A virus H3N2) using a clinical specimen, illustrating the ability of NTS to 248 detect and distinguish respiratory viruses. Together, our findings indicate that NTS is highly 249 suitable for the detection and variation monitoring of current COVID-19 epidemics, directly from 250 clinical samples with same-day turnaround of results. 251

At the time of this writing, the COVID-19 epidemic remains very severe. Accurate, rapid, and 252 comprehensive nucleic acid detection methods are needed to allow patients with suspected infection 253 to be isolated and treated as soon as possible, and to accurately confirm whether the patient is cured, 254 to prevent continued epidemic spread caused by misdiagnosis. The LoD of NTS was shown to be as 255 low as 10 copies/mL, rendering it 100-fold more sensitive than qPCR (e.g., some kits describe 256

LoDs of 1000 copies/mL) and thus likely to decrease the high false-negative rate plaguing current 257 detection methods. In addition, the detection of co-infection may allow the prevention of disease 258 progression from mild to severe or might be useful to inform clinical treatment. Overall, NTS 259 combines sensitivity, broad detection range, same-day rapid turnaround time, variation monitoring, 260 and low cost (compared with whole-genome sequencing), making it the most suitable method for 261 the detection of suspected viral infections that cannot be effectively diagnosed by other methods. 262

Moreover, the MinION, the smallest Oxford Nanopore sequencer, is smaller than a cellphone; when 263 equipped with a laptop computer for data processing, it thus allows rapid performance of NTS in 264 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 13 various environments with low equipment cost. For data analysis, cloud analysis may also be 265 introduced for high-throughput detection 17, 18 . 266

Several limitations of the current NTS method should be noted. Because the designed amplified 267 fragments are 300-950 bp in length, which constitute suitable lengths for detection by a nanopore 268 platform as nucleic acid fragments < 200 bp cannot be readily detected 19, 20 , thereby, the sensitivity 269 of NTS for detecting target COVID-19 fragments in highly degraded nucleic acids may be 270 hampered. Additionally, although the turnaround time of NTS is longer than that of qPCR or other 271 possible nucleic acid detection methods (e.g., SHERLOCK 12 ), 6-10 h is considered acceptable for 272 clinical use; moreover, NTS is already the fastest strategy based on sequencing methods to date and 273

can detect variations directly from clinical samples. Whereas the detection throughput of NTS is not 274 high at present, the NTS method can be integrated into widely used automated or semi-automated 275 platforms to improve the detection throughput in the future 21-23 . In addition, because PCR is 276 included in NTS, processes involving opening the lid of the PCR tubes may cause mutual 277 contamination between samples 24, 25 . However, this situation also is inevitable in current nucleic 278 acid detection methods (e.g., qPCR) or other nucleic acid detection schemes (e.g., SHERLOCK 11, 12 279 or toehold switch biosensor 9, 10 ) that also involve PCR. The introduction of integration systems or 280 sealed devices such as microfluidics may avoid this situation 26, 27 . At present, our processes of 281 sequencing data analysis and interpretation of results are not mature; nevertheless, as the number of 282 test samples increases, additional test results will be collected and the process continuously 283 optimized to obtain more accurate results. 284

Notably, the comparison of NTS and qPCR results indicated a high false-negative rate in the 285 latter. This result highlights the need for extreme vigilance, as patient misdiagnosis (including 286 patients admitted and discharged) will lead to spread of the epidemic and greater public health 287 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 14 threat. Suspected or negative results reported by the current qPCR methods should be subjected to a 288 more accurate method for secondary confirmation; for this, we consider NTS as the most 289 recommended solution currently available. The situation of co-infection, which has been reported in 290 our previous study 13 , also warrants continued attention. Based on the current centralized treatment 291 strategy, the lack of screening for multiple viruses may lead to large-scale cross-contamination and 292

confound clinical diagnosis and treatment. Alternatively, NTS represents and effective strategy that 293 can rapid and accurate distinguish SARS-CoV-2 and multiple respiratory viruses at both the species 294 and subtype level, and could be applied as a spot check in centralized clinics. Finally, the NTS 295 method for respiratory virus detection might be extended to detect more viruses and other pathogens 296 through the design of additional primer panels. N, and ORF10, was selected as a template to design a series of end-to-end primers. The region 303 encoding ORF1ab was selected as a template to design a nested primer for higher sensitivity 304 detection of SARS-CoV-2. All primers were designed using online primer-blast 305 (https://www.ncbi.nlm.nih.c/tools/primer-blast/) and the specificity of all primers was verified 306

against Homo sapiens, fungi, and bacteria. Finally, we downloaded and selected N, S, rdrp, and E 307 gene sequences of SARS-related viruses available at GenBank through January 1st, 2020 (accession 308 NC_045512). Multiple sequence alignment of SARS-CoV-2 against SARS-related viruses was 309 performed using Clustal W (version 1.83) for each gene individually and the alignment was used for 310 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint in-silico evaluation of specificity of the designed primers to SARS-CoV-2. All the specific primers 311 were collected to form the SARS-CoV-2 primer panel. 10, 100, 500, 1,000, and 3,000 copies/mL, with four replicates at each concentration. The NTS 326 libraries were prepared as described above and sequenced using MinION for 10 min, 30 min, 1 h, 2 327 h, and 4 h. The sequencing data were processed as described for virus identification. The LoD was 328 determined when the concentration of reads mapped to COVID-19 was significantly higher than 329 that for the negative control in 3/4 replicates. 330 331 NTS detection method. The targeted genes were amplified using the SARS-CoV-2 or 10 332 respiratory virus primer panel in a 20 μL reaction system with 5 μL total nucleic acid, 5 μL primer 333 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint (10 μM), and 10 μL 2× Phusion U Multiplex PCR Master Mix (Thermo Fisher, USA) 29, 30 . 334

Amplification was performed in a C1000 Thermocycler (Bio-Rad, USA) using the following 335 procedure: 1 cycle at 94 C for 3 min and 30 cycles at 95 C for 10 s, 55 C for 50 s, and 68 C for 336 5s, followed by a final elongation step at 68 C for 5 min. The product of the first-step was purified 337 with 0.8× AMpure beads (Beckman Coulter, USA) and eluted in 10 μL Tris-EDTA (TE) buffer. 338

Then, 5 μL eluate was used for second-step PCR with 5 μL barcoded primer (10 μM) and 10 μL 2× 339

Phusion U Multiplex PCR Master Mix. The barcode sequence was from the Nanopore PCR barcode 340 kit (EXP-PBC096; UK) and all primer oligos and full-length S and N gene fragments were 341 synthesized by Genscript (China). The products of second-step PCR from the different samples 342

were pooled with equal masses. TE buffer was assayed in each batch as a negative control. 343

Sequencing libraries were constructed using the 1D Ligation Kit (SQK-LSK109; Oxford Nanopore, 344 UK) and sequenced using Oxford Nanopore MinION or GridION. 345 346 Nanopore sequencing data processing. Basecalling and quality assessment for MinION 347 sequencing data were performed using high accuracy mode in Guppy (v. 3.1.5) software; for 348 GridION, the process was conducted using MinKNOW (v. 3.6.5) integrated in the instrument. orf1ab and E sites as the targets. Kit 1 is a cFDA-approved kit with two target sites used in this 499 study; kit 2 is a cFDA-approved kit with three target sites used in this study. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint

Distribution of novel coronavirus pneumonia

Clinical characteristics of 2019 novel coronavirus infection in China

No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint (which was not peer-reviewed) is

Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

Analysis of false-negative results for 2019 novel coronavirus nucleic acid test and related 428 countermeasures

Actionable diagnosis of neuroleptospirosis by next-generation sequencing

Zika virus evolution and spread in the Americas

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Toehold switches: de-novo-designed regulators of gene expression

Paper-based synthetic gene networks

Low-Cost Detection of Zika Virus Using Programmable Biomolecular Components

Nucleic acid detection with CRISPR-Cas13a/C2c2

A protocol for detection of COVID-19 using CRISPR diagnostics

Clinical diagnosis of 8274 samples with 2019-novel coronavirus in Wuhan

Analytical and clinical validation of a microbial cell-free DNA sequencing test for 443 infectious disease

Metagenomic Sequencing Detects Respiratory Pathogens in Hematopoietic Cellular 445 Transplant Patients

Capturing sequence diversity in metagenomes with comprehensive and scalable probe design

Nanopore metagenomics enables rapid clinical diagnosis of bacterial lower respiratory 449 infection

Rapid metagenomic identification of viral pathogens in clinical samples by real-time 451 nanopore sequencing analysis

Sequencing Platform. G3 (Bethesda)

High-Fidelity Nanopore Sequencing of Ultra-Short DNA Targets

Automation of PacBio SMRTbell NGS library preparation for bacterial genome sequencing

Automated Library Preparation for DNA Sequencing

Evaluation and optimisation of preparative semi-automated 461 electrophoresis systems for Illumina library preparation

Contamination in Low Microbial Biomass Microbiome Studies: Issues and 463 Recommendations

Laboratory diagnosis of Ebola virus disease

All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 17 subsequently. The taxonomy of each read was assigned according to the taxonomic information of 356 the mapped subject sequence. 357 358 Sequence correction and candidate mutation calling. Sequence correction was performed using 359 medaka 33 (v. 0.10.5), which is a tool to create a consensus sequence of nanopore sequencing data. 360For each target sequencing region, 30 consensus sequences were generated using medaka's default 361 settings through the medaka_consensus program. Subsequently, the consensus sequences were 362 aligned to the reference sequence of target sequencing regions using the multiple sequence 363 alignment tool ClustalW 34 (version 1.83). The variants within certainty regions (except sequence 364 homopolymeric regions and primer binding sites) 35 and with appropriate coverage (covered by at 365 least 90% consensus sequences and at least 50% uncorrected reads) were accepted as candidate 366 nucleotide mutations. 367

Interpretation of NTS results. The sequenced data were obtained at regular intervals after 369 sequencing, then filtered to obtain valid reads. For determining whether the target was positive, 370 interpretation was performed using the previous rule with modification [14] [15] [16] . In brief, if the read 371 matches the design fragment, the read will be counted. The mapping score was determined as 1, 0.4, 372 or 0 when the ratio of count number in the sample to the negative control of each target was > 10, 373 between 3-10, or < 3. The total mapping score of each target was summed and samples with > 2.4 374 total mapping score were defined as positive for SARS-CoV-2 infection; 1.2 to 2.4 total mapping 375 score indicated an inconclusive result, and < 1.2 total mapping score was considered to indicate 376 negative for infection. For determination of the other 10 kinds of common respiratory virus, a 377 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is 

All rights reserved. No reuse allowed without permission.the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 23 plasmid harboring an S gene was used as a positive control in NTS testing; a positive sample in the 517 kit served as a positive control in qPCR testing. NC: negative control. TE buffer was used as a 518 negative control in NTS testing; H2O in the kit served as a positive control in qPCR testing. Pos: 519 positive. Inc: inconclusive. Neg: Negative. 520 521 All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 26 528

All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint 27 530

All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint (which was not peer-reviewed) is . https://doi.org/10.1101/2020.03.04.20029538 doi: medRxiv preprint