key: cord-267426-3eu9umx5
authors: Yao, Hangping; Lu, Xiangyun; Chen, Qiong; Xu, Kaijin; Chen, Yu; Cheng, Linfang; Liu, Fumin; Wu, Zhigang; Wu, Haibo; Jin, Changzhong; Zheng, Min; Wu, Nanping; Jiang, Chao; Li, Lanjuan
title: Patient-derived mutations impact pathogenicity of SARS-CoV-2
date: 2020-04-19
journal: nan
DOI: 10.1101/2020.04.14.20060160
sha: 
doc_id: 267426
cord_uid: 3eu9umx5

The sudden outbreak of the severe acute respiratory syndrome-coronavirus (SARS-CoV-2) has spread globally with more than 1,300,000 patients diagnosed and a death toll of 70,000. Current genomic survey data suggest that single nucleotide variants (SNVs) are abundant. However, no mutation has been directly linked with functional changes in viral pathogenicity. Here we report functional characterizations of 11 patient-derived viral isolates, all of which have at least one mutation. Importantly, these viral isolates show significant variation in cytopathic effects and viral load, up to 270-fold differences, when infecting Vero-E6 cells. We observed intrapersonal variation and 6 different mutations in the spike glycoprotein (S protein), including 2 different SNVs that led to the same missense mutation. Therefore, we provide direct evidence that the SARS-CoV-2 has acquired mutations capable of substantially changing its pathogenicity.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2; previously referred to as 32 2019-nCoV), associated with the ongoing outbreak of atypical pneumonia, has already 33 caused a global pandemic, despite China's extensive systematic effort to contain the 34 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(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 3 outbreak. As of April 7, 2020, SARS-CoV-2 has infected more than 1.3 million people 35 around the world with a death toll of 70,000. The numbers are still increasing rapidly. 36 The estimate of the incubation period of SARS-CoV-2 (mean, 5.1 days; range, 4.5 to 5.8 37 days) (Lauer et al., 2020) is in line with those of other known human coronaviruses, such 38 as SARS (mean, 5 days; range, 2 to 14 days) (Varia et al., 2003) and MERS (mean, 5 to 7 39 days; range, 2 to 14 days) (Virlogeux et al., 2016) . The reproductive number of 40 SARS-CoV-2 is likely to be from 1.4 to 6.5, with a mean of 3.3 , which 41 is slightly higher than SARS, i.e. (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 April 19, 2020. . https://doi.org/10.1101/2020.04.14. can lead to devastating diseases, while HKU1, NL63, OC43 and 229E are related with 58 mild symptoms (Corman et al., 2018) . So far, no recombination events were detected (Yu, 59 2020), although this could be at least partially due to the fact that most viral isolates were 60 sequenced with short-reads platform. The transmembrane spike (S) glycoprotein mediates 61 viral entry into host cells through homotrimers protruding from the viral surface. The S 62 protein includes two domains: S1 for binding to the host cell receptor and S2 for fusion of 63 the viral and cellular membranes, respectively (Tortorici and Veesler, 2019) . Both 64 SARS-CoV-2 and SARS-CoV use the angiotensin converting enzyme 2 (ACE2) to enter 65 target cells (Walls et al., 2020) . ACE2 is expressed in human nasal epithelial cells and 66 lung, spermatogonia, leydig, sertoli, gastric, duodenal, and rectal epithelial cells (Wang 67 and Xu, 2020; Xiao et al., 2020; Zhao et al., 2020) . The receptor binding domain (RBD) 68 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 5 in the S protein is the most variable genomic part in the betacoronavirus group (Wu et al., 69 2020; Zhou et al., 2020), and some sites of S protein might be subjected to positive 70 selection (Lv et al., 2020) . Despite the abundant variability of SARS-CoV-2, one key 71 question remains as to whether these mutations have any real functional impact on the 72 pathogenicity of SARS-CoV-2. This is crucial in our understanding of the viral infectious 73 mechanisms and dictates the strategy of drug and vaccine development in preparation for 74 the next stage of the pandemic. 75 To address this, we characterized 11 SARS-CoV-2 viral isolates from patients (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Wuhan before they were diagnosed, and another 5 had close contact with people who 95 lived in Wuhan, and the remaining person had contact with people who were COVID-19 96 victims. Notably, patients ZJU-4, -5, -9 attended the same business conference where a 97 few colleagues from Wuhan were present. These patients therefore constitute 1st and 2nd 98 generations of the viral victims based on their epidemiological history. The 11 patients 99 include 8 males and 3 females, with ages ranging from 4 months to 71 years old. There 100 are no clear criteria in selecting these patients other than the fact that they were all 101 admitted into Zhejiang University-affiliated hospitals in Hangzhou. All except one of the 102 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. To assess the mutational spectrum of these 11 viral isolates, ultra-deep sequencing of the 112 isolated viral genomic RNA was performed on the Illumina Novaseq 6000 platform, 113 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 8 generating on average 245 million post-cleaning reads/67.16 Gb per sample (Table S1;   114 average coverage exceeding 2,000,000 X). This extraordinary depth is partially due to the 115 small genome of the SARS-CoV-2, which enables us to identify mutations with high 116 confidence. Moreover, in cases where the viral populations are not homogenous, the 117 depth could help us to characterize alleles with very low frequency. 118 In total, 33 mutations were identified (including 10 mutations observed in 119 mixed-populations), and 19 of these mutations were novel, according to the comparison 120 with 1111 genomic sequences available at GISAID on 3/24/2020 ( Fig. 1 , S1, and S2). 121 Specifically, G11083T and G26144T were found in ZJU-1, and both of these mutations 122 are known as founding mutations for a large group of viruses (Capobianchi et al., 2020) . 123 C8782T and T28144C were found in two of our viral isolates, ZJU-2 and ZJU-8, and 124 these two are known as the founding mutations for another large group of viral isolates 125 (Capobianchi et al., 2020) . Interestingly, mutation T22303G was found in five viral 126 isolates (ZJU-2, -5, -9, -10, and -11) and ZJU-5 and ZJU-9 were exposed to the same 127 potential source of infection during a business conference (Table 1) . Previously, only one 128 viral isolate identified in Australia had the T22303G mutation. Strikingly, the viral isolate 129 from patient ZJU-4, who attended the same conference as ZJU-5 and ZJU-9, has a novel 130 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 9 mutation, A22301C, which causes the same missense mutation at the protein level 131 (S247R in the S protein) as T22303G, mutating the 1st instead of the 3rd position in the 132 respective codon. Observations of these two single nucleotide variants can only be 133 coincidental, albeit very unexpected. Finally, the ZJU-11 has 4 mutations in the ORF7b 134 gene, 3 of which are consecutive and introduce 2 mutations at the protein level. 135 Di-nucleotide and Tri-nucleotide mutations are of course rarer than SNV, but not 136 exponentially so, according to previous mutational accumulation studies in prokaryotes 137 (Lynch, 2007) . 138 It is important to note that while the sequence data deposited in GISAID are very (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 April 19, 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 April 19, 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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 13 downloaded from GISAID. The 1000-times bootstrapped maximum likelihood tree was 179 constructed to demonstrate the phylogenetic context of the 11 viral isolates. Major and 180 minor clusters were color-coded and denoted as shown in the "colored ranges" inset box. (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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 14 monophyletic in our analysis ( Fig. 2A and S3 ). However, a distinct monophyletic 195 subclade of 92 sequences within the ORF8-L84S cluster can be observed, mainly 196 composed of viral sequences isolated from Seattle, USA ( Fig (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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 16 sequences and Taijima's D is -2.8874 with a nucleotide diversity (π) of 0.000641 (p < 229 0.05 according to simulations performed in (Tajima, 1989) , indicating that the 230 SARS-CoV-2 genome has an excess of low-frequency alleles due to recent population 231 expansions, consistent with the repeated bottlenecking events during viral infections. (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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 18 ZJU-11, decreased much faster than the others ( Fig. 3A and B) . At 48 hours P.I., we 263 observed small decreases for all viral isolates except for ZJU-10 and ZJU-11, both of 264 which had presumably already plateaued at 24 hours P.I. (Fig. 3A and B) . Notably, at 24 265 hours P.I., ZJU-2 and ZJU-8, members of the ORF-8-L84S cluster (majority of USA 266 WA-Seattle isolates are in this group), showed considerably lower viral loads (Fig. 4A) . 267 On the other hand, ZJU-1, which clusters with the S-D614G clade (mostly found in 268 Europe), has a viral load 19 times (2 4.25 ) higher than ZJU-2 and ZJU-8 (Fig. 4A) . In 269 addition, a near 270-fold difference (2 8.09 ) in viral load was observed between ZJU-10 270 and ZJU-2 at 24 hours P.I. (Fig. 4A ). These differences became statistically significant at their genomes, exhibit a significant variation of viral load when infecting Vero-E6 cells. 274 We next examined whether a higher viral load leads to more cell death (Fig. 4B) . 275 When examining these cell lines under a microscope at 48 hours and 72 hours P.I., the 276 CPE, or the cell death rate, are highly consistent with the viral load data ( Fig. 4C and S5B, 277 C, and D; at 48 hours P.I., Ct vs CPE, R = -0.72, p = 0.015), indicating that a higher viral 278 load leads to a higher cell death ratio. Note that the Ct numbers are negatively correlated 279 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. (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 April 19, 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 April 19, 2020. Pair-wise p-values were calculated between isolates using the t-test and adjusted p-values 298 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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint removed. Although the Vero-E6 cell line was not derived from human, the ACE2 protein 316 of the Vero-E6 cell line is highly similar to that of Human (Fig. S6 ) and we provided 317 direct evidence that the SARS-CoV-2 can infect the cell line (Fig. 2B) . 318 Several findings stand out in our study: 1. A diverse collection of mutations was 319 identified in the 11 viral isolates, including two sets of founding mutations for two major 320 clusters of viruses currently infecting the world population. In addition, 19 of the 31 321 identified mutations are novel, despite the relatively early sampling dates, indicating that 322 the true diversity of the viral strains is still largely underappreciated; 2. remarkably, the 323 T22303G and A22301C mutations result in the same S247R mutation in the S protein 324 ( Fig. 1 and S1 ), mapping to the existing structure revealed that this residue is located in a 325 flexible loop region within the N-terminal domain of the S1 subunit of S protein, 326 although the exact position of S247 could not be determined (Fig. S7, red arch) . While 327 the N-terminal domain is not directly involved with binding to ACE2 (Walls et al., 2020) 328 we note that this domain is positioned right next to the C-terminal domain, which binds to 329 ACE2. Interestingly, the T22303G mutation was observed in 5 viral isolates, albeit in 330 different proportions, indicating that this specific mutation was already present in the 331 early days of pandemic, and probably in a significant number of people of Wuhan, 332 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 24 despite the fact that it is still largely missing from current GISAID collection. This could 333 be due to the founding effect of mutations, in which case the T22303G mutation was not 334 transmitted out of the China during the early days; 3. The tri-nucleotide mutation in 335 ZJU-11 is unexpected; we note that this specific viral isolate is quite potent in our viral 336 load and CPE assay, and its patient remained positive for an astounding period of 45 days 337 and was only recently discharged from the hospital (Table 1) . Investigating the functional 338 impact of this tri-nucleotide mutation would be highly interesting. We note that in the 339 current database, another trinucleotide mutation (G28881A, G2882A and G28883C) has 340 been identified, which also results in two missense mutations at the protein level (Fig. S8) . 341 It leads to a cluster of more than 300 viral strains as of the time of writing this article, and 342 its mutational impact on the viral pathogenicity would be worth investigating. Finally, in 343 contrary to the recent report that a viable viral isolate could not be obtained from stool 344 samples, three of our viral isolated were extracted from stool samples, indicating that the 345 SARS-CoV-2 is capable of replicating in stool samples (Woelfel et al., 2020) . 346 In short, our study provides direct evidence that mutations currently occurring in the 347 SARS-CoV-2 genome have the functional potential to impact the viral pathogenicity. 348 Therefore, viral surveillance should be also performed at the cellular level when possible, 349 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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv 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 preprint this version posted April 19, 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 April 19, 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 April 19, 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 April 19, 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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint Sequencing library construction 478 The total RNA in each deactivated viral sample was extracted using a viral RNA mini kit 479 (Qiagen, Germany). The sequencing library was constructed using the Total RNA-Seq (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Sequence data processing, de novo assembling, and mutation identifications 506 Sequencing data was generated from Novaseq 6000 and first filtered of low quality and References of SARS-CoV-2 were downloading from NCBI on date February 28th, 2020. 510 Further, mapping reads that longer than 100nt were extracted for de novo assemby by 511 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. for these samples. After assembling, contigs was blasted to nt database (20190301) to 514 confirm their origins, and only contigs belonging to coronavirus were retained for base 515 correction. Next, filtering reads of each sample were mapped back to retained assembled 516 contigs and bam-readcount was applied (--min-mapping-quality=5, other parameter was 517 set default) to calculate the base frequency of every post of each assemble contigs. 518 Meanwhile, Haplotypecaller of gatk was applied to call snp/indel based on the assembled 519 contigs with reads quality higher than 20. Finally, bam files were inspected in igv 520 manually to verify each mutation based on the number of reads mapped, the balance 521 between reads mapped to plus and minus strands of the reference genome, and the 522 relative positions of the mutations on these reads. 523 524 Phylogenetic analysis 525 We acquired 725 high quality and high coverage SARS-CoV-2 genomes from GISAID 526 (downloaded on 3/21/2020), including the Yunnan RaTG13 viral strain and the 527 Guangdong pangolin viral strain as the outgroup. We aligned the 736 genomic sequences 528 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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv 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 preprint this version posted April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv 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 preprint this version posted April 19, 2020. . https://doi.org/10.1101/2020.04.14. was in place to represent to the potential flexible loop conformation for (C) and (D). Also 586 note that the protein complex is trimeric, but only one of the three mutations was labeled. 587 The 3D structure of the S protein was visualized and downloaded from 588 https://www.rcsb.org/3d-view/6VSB/1. 589 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv 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 preprint this version posted April 19, 2020. . https://doi.org/10.1101/2020.04.14. 

Carrier Transmission of COVID-19

SPAdes: A new genome assembly algorithm and its 613 applications to single-cell sequencing

Dynamically modeling SARS and 615 other newly emerging respiratory illnesses: Past, present, and future

The control of the false discovery rate in multiple testing under 617 dependency

A 619 Trial of Lopinavir-Ritonavir in Adults Hospitalized with Severe Covid-19

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

trimAl: A tool for automated 621 alignment trimming in large-scale phylogenetic analyses

Molecular characterization of SARS-CoV-2 from the first case of 624 COVID-19 in Italy

COVID-19 clinical diagnosis and management guideline issued 626 by National Health Commission of China

Hosts and Sources of Endemic Human 628

Advances in Virus Research

Aerosol and Surface Stability of

SARS-CoV-2 as Compared with SARS-CoV-1

Molecular Evolution of Human Coronavirus 633

Clinical characteristics of 2019 novel coronavirus infection in China

637 TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Clinical characteristics of 24 asymptomatic infections with COVID-19 screened among close contacts in 641

MAFFT Multiple Sequence Alignment Software Version 7: 643 Improvements in Performance and Usability

MEGA7: Molecular Evolutionary Genetics Analysis 645

Version 7.0 for Bigger Datasets

The Incubation Period of Coronavirus Disease

Reported Confirmed Cases: Estimation and Application

Interactive Tree of Life v2: Online annotation and display of phylogenetic 650 trees made easy

Fast and accurate short read alignment with Burrows-Wheeler transform

Modeling the spread of Middle East respiratory

Transmission dynamics and control of severe acute respiratory syndrome

The reproductive number of COVID-19 is 659 higher compared to SARS coronavirus

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and 662 receptor binding

Comparative genomic analysis revealed specific 664 mutation pattern between human coronavirus SARS-CoV-2 and Bat-SARSr-CoV RaTG13

The Origins of Genome Architecture

IQ-TREE: A fast and effective 668 stochastic algorithm for estimating maximum-likelihood phylogenies

RBD 670 mutations from circulating SARS-CoV-2 strains enhance the structure stability and infectivity of the spike

On the Analysis of

Intrahost and Interhost Viral Populations: Human Cytomegalovirus as a Case Study of Pitfalls and 674

Acute Respiratory Syndrome Coronavirus Replication Is Severely Impaired by MG132 due to 677

Proteasome-Independent Inhibition of M-Calpain

Statistical method for testing the neutral mutation hypothesis by DNA polymorphism

Structural insights into coronavirus entry

Investigation 682 of a nosocomial outbreak of severe acute respiratory syndrome (SARS) in Toronto

Comparison of incubation period 685 distribution of human infections with MERS-CoV in South Korea and Saudi Arabia

Antigenicity of the SARS-CoV-2 Spike Glycoprotein

scRNA-seq profiling of human testes reveals the presence of ACE2 receptor, 689 a target for SARS-CoV-2 infection, in spermatogonia

A human monoclonal 1 antibody blocking 692 SARS-CoV-2 infection

Clinical presentation and virological assessment of hospitalized 695 cases of coronavirus disease 2019 in a travel-associated transmission cluster

A new coronavirus associated with human respiratory disease in China

Evidence for gastrointestinal infection 700 of SARS-CoV-2

Decoding evolution and transmissions of novel pneumonia coronavirus (SARS-CoV-2)

using the whole genomic data Comparative analyses of the chloroplast genome in carnivorous plants View 703 project

Single-cell RNA expression profiling 705

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. (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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 43 syndrome coronavirus in Saudi Arabia. Stat. Methods Med. Res. 27, 1968 Res. 27, -1978 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. (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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint 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 April 19, 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 April 19, 2020. . https://doi.org/10.1101/2020.04.14.20060160 doi: medRxiv preprint