key: cord-0963820-hdgvfsxa
authors: Telwatte, Sushama; Kumar, Nitasha; Vallejo-Gracia, Albert; Kumar, G. Renuka; Lu, Chuanyi M.; Ott, Melanie; Wong, Joseph K.; Yukl, Steven A.
title: Novel RT-ddPCR assays for simultaneous quantification of multiple noncoding and coding regions of SARS-CoV-2 RNA
date: 2021-03-02
journal: J Virol Methods
DOI: 10.1016/j.jviromet.2021.114115
sha: 9a447065c708ea505bea18d44e0089afdc5e9a79
doc_id: 963820
cord_uid: hdgvfsxa

A hallmark of coronavirus transcription is the generation of negative-sense RNA intermediates that serve as the templates for the synthesis of positive-sense genomic RNA (gRNA) and an array of subgenomic mRNAs (sgRNAs) encompassing sequences arising from discontinuous transcription. Existing PCR-based diagnostic assays for SAR-CoV-2 are qualitative or semi-quantitative and do not provide the resolution needed to assess the complex transcription dynamics of SARS-CoV-2 over the course of infection. We developed and validated a novel panel of sensitive, quantitative RT-ddPCR assays designed to target regions spanning the genome of SARS-CoV-2. Our assays target untranslated regions (5’, 3’) as well as different coding regions, including non-structural genes that are only found in full length (genomic) RNA and structural genes that are found in genomic as well as different subgenomic RNAs. Application of these assays to clinically relevant samples will enhance our understanding of SARS-CoV-2 gene expression and may also inform the development of improved diagnostic tools and therapeutics.

lesions 20 or indirectly inhibit immune responses 21 . Incorporation of 5'UTR sequences into the capped subgenomic mRNA templates of SARS-CoV may confer resistance to cleavage by viral nsp1 protein 22 , which typically inhibits host gene expression by degradation of host mRNA [23] [24] [25] . For positive-sense RNA viruses, sgRNAs act as messengers for expression of structural proteins or proteins related to pathogenesis and can regulate the transition between translation and virion production 26 . The various roles of sgRNAs in SARS-CoV-2 infection and pathogenesis remain to be elucidated, but the rapid accumulation and persistence of sgRNAs following infection may also contribute to disease progression. Existing PCR-based diagnostic assays for SAR-CoV-2, which are interpreted in a qualitative or semi-quantitative manner (positive, negative or indeterminate) and target only 1-2 viral regions, do not account for possible variation in RNA copy numbers due to subgenomic transcription. For example, the more 5' genes in ORF1a and ORF1b are excluded from subgenomic transcripts and therefore may be present at lower levels than genes in ORF 2-10, which are present in subgenomic as well as genomic transcripts. Among the 'body' genes found in sgRNAs, those at the 3' end [ORF [9] [10] would be expected to present in all sgRNAs and therefore might be present at higher copy numbers than more 5' body genes, such as ORF2, which is only present in one type of subgenomic transcript. Therefore, RNA levels of a given gene may depend on the degree to which it is transcribed as various sgRNAs 9 , the degree to which the sample includes virion or cell-associated RNAs, and the degree to which sgRNAs may persist in double membraned vesicles even after replication has ceased 27, 28 . Molecular assays that can quantify different genes present in sgRNA and/or gRNA species may prove useful for improving clinical diagnostic tests and for research in understanding how the regulation of viral gene expression contributes to clinical disease. To help investigate these questions, we devised a novel panel of seven ddPCR-based assays that target various conserved regions of SARS-CoV-2 RNA, including the 5' and 3' untranslated regions, non-structural genes that are only found in full length (genomic) RNA, and structural genes that are also contained in different sgRNAs ( Fig.1 and Table 1 ). [29] [30] [31] , we designed a primer/probe set to target the short, highly-conserved 'polybasic cleavage site' ('S-PBCS') of SARS-CoV-2, which is functionally cleaved to yield the S1 and S2 subunits 32 , in a manner similar to the hemagglutinin (HA) protein of avian influenza viruses (AIVs) 33 . In AIVs, the insertion or substitution of basic amino acids at the HA cleavage site is associated with enhanced pathogenicity 34, 35 .The SARS-CoV-2 PBCS allows effective cleavage by host furin and other proteases 5 , and may potentially enhance its infectivity in humans and distinguish it from related animal coronaviruses 4, 5, 36 . Elucidating the granular detail of SARS-CoV-2 transcription could help us to understand how the virus replicates and how it may evade human immune defenses. Detailed mapping of the expressed viral transcripts across times and cell types is essential for further studies of viral gene expression, mechanisms of replication, and probing host-viral interactions involved in pathogenicity.

Multiple primer and probe sets were designed to target various regions of SAR-CoV- HcoV-HKU1 (NC_006577.2), were included in the alignment to exclude primer sets with significant overlap with non-SARS-CoV-2 sequences. For each region, we selected at least one primer/probe set (and where possible, an alternative set) that aligned to all SARS-CoV-2 isolates but had 1 or more mismatch with SARS-CoV and greater than 5 mismatches with MERS-CoV, HCoV-229E, HCoV-NL63, HCoV-OC43, and HCoV-HKU1 (Table 1) . A sequence similarity analysis using Basic Local Alignment Search Tool (BLAST) 38 found no significant similarity in any primer or probe to human sequences. RNAs, for which the combination of poly-dT plus random hexamers may reduce bias towards reverse transcription of any one region (as can be seen with specific reverse primers), the 5' J o u r n a l P r e -p r o o f end (as would be expected with random hexamers), or the 3' end (as would be expected with poly-dT).

IVT RNA standards were added to RT reactions at inputs of 1, 10, 10 2 , 10 3 , and 10 4 copies per 5 µL (2 replicate RT reactions for each input). RT reactions were performed in a conventional thermocycler at 25.0°C for 10 min, 50.0°C for 50 min, followed by an inactivation step at 85.0°C for 5 min. Undiluted RT product (5 µL) was added to ddPCR reactions (total volume of 20 µL) and ddPCR was performed as described in Section 2.2. Alternative primer/probe sets for a given region were compared head-to-head using this approach. Based on performance of each primer/probe set using plasmid DNA and IVT RNA, one primer/probe set for each region was selected for further testing.

To determine the robustness of our approach, in addition to testing each assay with varying RNA copy inputs (each with two replicate RT reactions per input and replicate ddPCR wells for each RT), we performed repeat, independent experiments using the same parameters to confirm each assay's efficiency and sensitivity (n=4 for N-ORF9, CDC_N1, and CDC_N2; n=3 for 5'UTR, 3'UTR; and n= 2 for all others). No data were excluded as outliers. were added to RT reactions to achieve expected inputs of 1 to 70,000 copies per 5 µL RT (the input into each ddPCR well). RT reactions were performed as above, with random hexamers and poly-dT, except that the total volume of the RT was scaled up so that two replicate 5 µL aliquots of cDNA could then be used to test each assay in parallel using replicate 20 µL ddPCR reactions (as described in Sections 2.2 and 2.3) containing primers/probe specific for a given region.

To determine whether a 1-step dd-RT-PCR approach could be adopted for each SARS-CoV-2 assay, a total of 1000 RNA copies/final ddPCR well (supernatant standard) were added to 20 µL reactions containing: 1x 1-step RT-ddPCR Supermix (Bio-Rad), 1 mM manganese acetate, 900 nM of primers, and 250 nM of probe. Droplets were amplified using a Mastercycler® nexus (Eppendorf, Hamburg, Germany) with the following cycling conditions: a reverse transcription reaction at 65°C for 30 min, inactivation at 95°C for 5 min, followed by 40 cycles of 30 s at 94°C and 60°C for 60 s, and a final droplet cure step of 10 min at 98°C.

Droplets were read and analyzed using the QuantaSoft software in the absolute quantification mode and compared to samples run in parallel with 1000 copies/final ddPCR well using the 2-step approach, in which the RT reaction was performed separately and undiluted RT product (5 µL from the same RT reaction) was added to ddPCR reactions (total volume of 20 µL).

Further validations were performed to determine each assay's sensitivity to inhibition was added to each 20 L ddPCR reaction and replicate ddPCR reactions were performed for each assay.

To investigate the variations in gene expression in clinical samples and determine whether our RT-ddPCR assays correlate with a clinical test, we obtained unused nucleic acid J o u r n a l P r e -p r o o f 3. Results

Two primer/probe sets were designed for each region (indicated in Fig. 1 ; 'Primer locations') except the spike protein polybasic cleavage site, for which only one primer/probe set was designed. To evaluate the performance of each primer/probe set at the PCR stage, each set was first tested on plasmid DNA. Since no commercially available plasmid contains the whole SARS-CoV-2 genome, and construction of such a plasmid is technically challenging (due to the 30kb length) and subject to higher biosafety restrictions, we constructed or purchased plasmids containing individual genes or regions. For each plasmid, the DNA concentration was measured by UV spectroscopy (NanoDrop) and the number of molecules (expected copies) was calculated using the molecular weight.

Each primer/probe set was assessed for detection limit, dynamic range, linearity, and efficiency by measuring the absolute number of copies detected using droplet digital PCR (ddPCR) from expected inputs of serially diluted plasmid DNA. All primer/probes sets could detect as few as 1-10 copies and were linear over at least 3 orders of magnitude (R 2 >0.99 for all; Fig. 2 ). Assay efficiencies (measured by the slope) ranged from 0.67 ("N-ORF9_8") to 1.1 ("M-ORF5"). One primer/probe set for each region was selected for further study (Table 2; rejected primer/probe sets are listed in Table S1 ) based on the overall efficiency ( Fig. 2) , separation between the positive and negative droplets (signal to noise ratio; Fig. S1 ), and specificity (Table S2) . For the chosen assays, no positive droplets were detected with water or DNA from peripheral mononuclear blood cells (PBMC) from uninfected blood donors.

J o u r n a l P r e -p r o o f regions were quantified by UV spectroscopy and diluted (expected copies) to test the absolute number of copies detected by each primer/probe set using duplicate ddPCR reactions (measured copies). Two primer/probe sets were tested for each region except the S-PBCS.

One primer/probe set from each region (indicated by coloured symbol) was selected for subsequent experiments.

J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f

Selected primer/probe sets for each region were tested using standards prepared from in vitro transcribed (IVT) RNA generated from the designed plasmids (5'UTR, Main Proteinase, RDRP, S, M, N and 3'UTR; Fig. 3 and Table 2 ). The expected copy numbers were calculated using the RNA concentration (as measured by UV spectroscopy [NanoDrop] and confirmed by the Agilent Bioanalyzer) and the molecular weight. Using RT-ddPCR, all assays could detect as few as 10 copies of RNA and demonstrated linearity over 3-4 orders of magnitude (R 2 >0.999 for all; Fig. 3 ). The efficiencies for detecting IVT RNA standards, which ranged from 0.18 (for Main Proteinase) to 0.96 (S-PBCS), were more variable than those observed for plasmid DNA. No amplification was detected in 'No RT' control reactions containing 10,000 IVT RNA copies/well, confirming the absence of any contaminating plasmid DNA. However, it is worth noting that none of these IVT standards were polyadenylated (so they should not be reverse-transcribed by poly-dT) and some of the standards were very short (<300 base pairs), which would likely limit the efficiency with which they were reverse transcribed by random hexamers. In addition, some of the measured differences in efficiency could reflect actual differences in the copy numbers present in the various IVT standards, which are difficult to determine precisely. 

To determine the non-specific reactivity of oligonucleotides (false positive rate) for each primer/probe set, we performed a median of 26 [range [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] 'no template' controls (NTC). These reactions were performed with both water (water NTC) and DNA or RNA isolated from SARS-CoV-2-negative donor PBMC (DNA/RNA NTC) (Table S2 ). Except for one experiment using IVT RNA, where a total of three droplets were detected across duplicate NTC wells containing donor PBMC tested for Main Proteinase-NSP5, no other false positives were observed.

Our assay panel included new primers/probes for the nucleocapsid (N-ORF9), which is targeted by existing diagnostic real-time PCR assays. We compared the performance of our 'N-ORF9' primers/probe to the primers/probes from the U.S. Center for Disease Control assays for the nucleocapsid (CDC-N1 and CDC-N2) 40 using ddPCR. The N-ORF9 assay efficiency was similar to that of CDC-N1 and CDC-N2 for plasmid DNA, in between that of CDC-N1 and CDC-N2 for IVT RNA, and similar to CDC-N1 for the supernatant standard ( Fig.   2-4 ).

In addition, we compared our primers/probes for the RdRp to published primers/probes for the "IP2" assay 41 (which targets ORF1a) and "E-Sarbeco" 42 assay (which targets the E gene) using RT-ddPCR and the supernatant standard ( Fig. 5 ; Table 3 ). The IP2 (ORF1a) assay efficiency was 1.11, compared to 1.20-1.28 for our RdRp (ORF1b) and 1.36 for our main protease (ORF1a) assays (Fig. 4-5) . The E-Sarbeco [ORF4] assay efficiency (1.08) was similar to the IP2, but may have been lower efficiency than our primer/probe sets targeting transcription using random hexamers and poly-dT, followed by droplet encapsulation and ddPCR). We observed that the 2-step approach was similar to or outperformed the 1-step dd-RT-PCR method in terms of detection at the same input (1000 copies) for 5'UTR, RdRp, S-PBCS, M-ORF5, and 3'UTR (Fig. S2) . The 1-step method enabled higher detection of ProNSP5, N-ORF9 and CDC_N2 relative to the 2-step approach, but the 1-step showed more false positives (N-ORF9 and CDC_N2) or poorer signal to noise ratio (ProNSP5). Overall, the performance of the SARS-CoV-2 panel of ddPCR assays was superior using a 2-step approach compared to a 1-step dd-RT-PCR approach.

J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f

Our validation studies included SARS-CoV-2 RNA inputs of as few as 1 copy per ddPCR reaction (Fig. 2-3 ). We estimated the lower limit of detection (LLOD) for each assay in our panel based on data for all replicates tested at 10 copy and 1 copy inputs (Table S3) . At 10 copies, all of our assays detected SARS-CoV-2 in ≥85.7% of tests (range= 85.7-100%). At 1 copy input, our assays detected SARS-CoV-2 in ≥25% of tests (range=25-88%),

underscoring the high sensitivity of our assays.

Next, we assessed the efficiencies of the primer/probe sets in our panel ( 

To 

The 2019 SARS-CoV-2 outbreak has heralded the development of an array of diagnostic molecular tools to study this novel coronavirus. However, currently described PCRbased diagnostic assays are qualitative or semi-quantitative, are limited to the simultaneous detection of one or two regions (typically RdRp and N), and do not account for variation in gene copy numbers due to subgenomic transcription. Here, we report a panel of new primer/probe sets that span the SARS-CoV-2 genome and target important nongenic regions, non-structural genes found only in genomic RNA, and structural genes that are also found in different subgenomic RNAs.

We used these new primers/probes for RT-ddPCR rather than qRT-PCR because ddPCR provides absolute quantification (does not require an external calibrator), tends to tolerate sequence mismatches in primer/probe sequences better than qRT-PCR, and may be more precise at low copies, while providing similar sensitivity and reproducibility 39, 44 . During J o u r n a l P r e -p r o o f validation of these assays with multiple different standards, we sometimes found that the efficiency of the same assay varied across different standards. The greatest variation was observed using in vitro transcribed standards, whose varying length and lack of a poly-A tail may affect RT efficiency (Fig. 3) . Together, these observations may reflect differences in the nature of the standards (DNA, short in vitro transcribed non-polyadenylated RNA, or supernatant RNA) as well as the difficulty in determining the exact number of copies in an external standard; the latter issue highlights a major advantage of the absolute quantification provided by ddPCR.

On all standards tested, the seven SARS-CoV-2 RT-ddPCR assays were extremely sensitive (down to 1-10 copies) and linear over 3-4 orders of magnitude. All seven assays showed no inhibition by up to 500,000 cell equivalents of RNA per ddPCR well, suggesting that these assays could be extremely useful for SARS-CoV-2 research. While most existing clinical assays for SARS-CoV-2 use qPCR because it is less expensive and may have fewer false positives than ddPCR, it is likely that the primer/probe sets described here would also work well in qPCR assays for research or clinical testing.

The utility of assays that target multiple genomic regions is supported by studies demonstrating loss in sensitivity of published assays owing to mutations that could affect primer annealing. For instance, a recent study found that 34.38% (11, 627) 56 . Therefore, sensitive RT-ddPCR assays such as those described in this study could be of great utility in studying the course of infection two or more weeks after the resolution of acute symptoms.

Another advantage of the approach described here is that it permits a single sample to be simultaneously assayed for multiple targets, which may increase sensitivity and specificity compared to measuring only 1-2 targets while helping to delineate the levels of various SARS-CoV-2 coding regions in infected patient samples from cross-sectional or longitudinal studies.

As such, this panel of RT-ddPCR assays can be applied to a diverse range of clinically relevant samples in which SARS-CoV-2 RNA may be present in low or high abundance.

Using both the supernatant virus standard and clinical samples from the nasopharynx, we tended to observe higher copy numbers for targets at the 3' end of the genome (N, 3'UTR) compared to the 5' end (5'UTR, main protease) (Fig. 4, 6, 7 and S2 ). This discrepancy is not explained by differences in PCR efficiency, since the efficiency of the N assay on plasmid DNA was actually lower than that of assays for the 5'UTR or main protease. It is possible that reverse transcription is more efficient for assays at the 3' end (perhaps due to more efficient reverse transcription from the poly-dT), although random hexamers should bias towards the 5' end and the combination has been used to prevent bias towards either the 5' or 3' end of J o u r n a l P r e -p r o o f the 9.6kb genome of HIV-1 39,57 . Moreover, the supernatant levels of 3' RNA regions also tended to be higher than those of 5' regions when measured using a "1-step" approach in which reverse transcription was performed using specific reverse primers. It seems likely that the 3' assays measure higher copies because they are detecting subgenomic RNAs, which are reported to be excluded from virions 58 The clinical implications of SARS-CoV-2 subgenomic RNA transcription are currently unknown. The synthesis of subgenomic RNAs is a common strategy employed by positivesense RNA viruses to transcribe their 3' proximal genes that encode products essential for particle formation and pathogenesis [59] [60] [61] . In coronaviruses such as mouse hepatitis virus considerably. Furthermore, our comparison of viral loads obtained by RT-ddPCR and qRT-PCR demonstrates the strong correlation between data obtained from these two platforms and the minimal RNA input required to yield robust data using our RT-ddPCR assays.

Limitations of this study should be acknowledged. In order to test our primer/probe sets in parallel with published assays (total of 11 assays) in background RNA experiments ( Fig. 6) , we increased RT reaction volumes from 50-70 µL to 125 µL to accommodate the additional assays. In the absence of background RNA, the efficiency appeared to be higher in the 50-70µL RT reactions ( Fig. 4-5 , >100% efficiency for all assays) than the 125µL reactions ( Fig. 7 ; median efficiency=88% [range: 60-133%]). If the discrepancy is not due to a difference in the actual input of the standard, it is possible that larger reaction volumes lead to lower efficiency in reverse transcription. However, for application to patient samples, our core panel of 7 RT-ddPCR assays (5'UTR, Pro-NSP5, RdRp-NSP12, S-PBCS, M-ORF5, N-ORF9, and 3'UTR; Table 1 ) is sufficient to provide a detailed view of the transcription profile of SARS-CoV-2, so preparation of RT reactions >70µL will likely be unnecessary. 

We developed a panel of sensitive, quantitative RT-ddPCR-based SARS-CoV-2 assays that collectively span the genome and target nongenic and genic regions, including important enzymes transcribed only as genomic RNA and structural genes that are also transcribed as different subgenomic RNAs. These assays can serve as novel molecular tools to investigate SARS-CoV-2 infection, replication dynamics, and gene expression to better understand the viral dynamics and pathogenesis of SARS-CoV-2 over the course of infection. Future studies employing these assays will enhance our understanding of SARS-CoV-2 replication and transcription and may also inform the development of improved diagnostic tools and therapeutics.

The study authors had no subject contact or access to any personally-identifiable information (Category 4, IRB exempt). The data generated and/or analyzed during the current study are included in the manuscript or supplements. All authors have seen and approved the content and have contributed to the work. The authors declare no conflicts of interest. None of the content has been published or is under consideration for publication elsewhere.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. For each assay, number of wells in which at least one droplet was detected is shown.

*A total of 3 droplets were detected in two wells from the same assay.

J o u r n a l P r e -p r o o f 

A Novel Coronavirus from Patients with Pneumonia in China

Identification of Coronavirus Isolated from a Patient in Korea with COVID-19

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

A Novel Bat Coronavirus Closely Related to SARS-CoV-2

Contains Natural Insertions at the S1/S2 Cleavage Site of the Spike Protein

The proximal origin of SARS-CoV-2

An update on origin of SARS-CoV-2: Despite closest identity, bat (RaTG13) and Pangolin derived Coronaviruses varied in the critical binding site and O-linked glycan residues

Coronavirus biology and replication: implications for SARS-CoV-2

A Contemporary View of Coronavirus Transcription

The Architecture of SARS-CoV-2

Expression, purification, and characterization of SARS coronavirus RNA polymerase

Multiple enzymatic activities associated with severe acute respiratory syndrome coronavirus helicase

A Structural View of SARS-CoV-2 RNA Replication Machinery: RNA Synthesis, Proofreading and Final Capping

Continuous and Discontinuous RNA Synthesis in Coronaviruses

A unifying structural and functional model of the coronavirus replication organelle: Tracking down RNA synthesis

A new coronavirus associated with human respiratory disease in China

Minus-strand copies of replicating coronavirus mRNAs contain antileaders

Coronaviruses use discontinuous extension for synthesis of subgenome-length negative strands

Characterization and complete genome sequence of a novel coronavirus, coronavirus HKU1, from patients with pneumonia

Direct RNA nanopore sequencing of full-length coronavirus genomes provides novel insights into structural variants and enables modification analysis

The murine coronavirus nucleocapsid gene is a determinant of virulence

Severe acute respiratory syndrome coronavirus open reading frame (ORF) 3b, ORF 6, and nucleocapsid proteins function as interferon antagonists

SARS coronavirus nsp1 protein induces template-dependent endonucleolytic cleavage of mRNAs: viral mRNAs are resistant to nsp1-induced RNA cleavage

Severe acute respiratory syndrome coronavirus nsp1 suppresses host gene expression, including that of type I interferon, in infected cells

A two-pronged strategy to suppress host protein synthesis by SARS coronavirus Nsp1 protein

SARS-coronavirus replication/transcription complexes are membrane-protected and need a host factor for activity in vitro

SARS-CoV-2 genomic and subgenomic RNAs in diagnostic samples are not an indicator of active replication

Tracking Changes in SARS-CoV-2 Spike: Evidence that D614G Increases Infectivity of the COVID-19 Virus

First detection of SARS-CoV-2 spike protein N501 mutation in Italy in

Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein

Emergence of a highly pathogenic avian influenza virus from a low-pathogenic progenitor

Origin and molecular changes associated with emergence of a highly pathogenic H5N2 influenza virus in Mexico

Genetic Predisposition To Acquire a Polybasic Cleavage Site for Highly Pathogenic Avian Influenza Virus Hemagglutinin

The EMBL-EBI search and sequence analysis tools APIs in 2019

Gapped BLAST and PSI-BLAST: a new generation of protein database search programs

2019-Novel Coronavirus (2019-nCoV) Real-time RT-PCR Primers and Probes. National Center for Immunization and Respiratory Diseases (NCIRD), Division of Viral Diseases

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

Analytical sensitivity and efficiency comparisons of SARS-CoV-2 RT-qPCR primer-probe sets

Droplet digital PCR (ddPCR) vs quantitative real-time PCR (qPCR) approach for detection and quantification of Merkel cell polyomavirus (MCPyV) DNA

Multiple assays in a real-time RT-PCR SARS-CoV-2 panel can mitigate the risk of loss of sensitivity by new genomic variants during the COVID-19 outbreak

Comparing the analytical performance of three SARS-CoV-2 molecular diagnostic assays

Performance Assessment of SARS-CoV-2 PCR Assays Developed by WHO Referral Laboratories

Detection of SARS-CoV-2 RNA by multiplex RT-qPCR

ddPCR: a more accurate tool for SARS-CoV-2 detection in low viral load specimens

Profile of RT-PCR for SARS-CoV-2: a preliminary study from 56 COVID-19 patients

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang province, China

Positive RT-PCR Test Results in Patients Recovered From COVID-19

Testing Dilemmas: Post negative, positive SARS-CoV-2 RT-PCR -is it a reinfection

Kinetics of SARS-CoV-2 positivity of infected and recovered patients from a single center

Prolonged SARS-CoV-2 RNA shedding: Not a rare phenomenon

Gut and blood differ in constitutive blocks to HIV transcription, suggesting tissue-specific differences in the mechanisms that govern HIV latency

Virological assessment of hospitalized patients with COVID-2019

A positive-strand RNA virus with three very different subgenomic RNA promoters

Nidovirus transcription: how to make sense

Nidovirales: evolving the largest RNA virus genome

Subgenomic negative-strand RNA function during mouse hepatitis virus infection

Expanded subgenomic mRNA transcriptome and coding capacity of a nidovirus

An overview of technical considerations when using quantitative real-time PCR analysis of gene expression in human exercise research

Variables influencing the efficiency and interpretation of reverse transcription quantitative PCR (RT-qPCR): An empirical study using Bacteriophage MS2

Detection limits of several commercial reverse transcriptase enzymes: impact on the low-and highabundance transcript levels assessed by quantitative RT-PCR