Recombinase Polymerase Amplification Combined with Fluorescence Immunochromatography Assay for On-Site and Ultrasensitive Detection of SARS-CoV-2

This study established a portable and ultrasensitive detection method based on recombinase polymerase amplification (RPA) combined with high-sensitivity multilayer quantum dot (MQD)-based immunochromatographic assay (ICA) to detect the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The RPA-MQD-based ICA method is reported for the first time and has the following advantages: (i) RPA is free from the constraints of instruments and can be promoted in point-of-care testing (POCT) scenarios, (ii) fluorescence ICA enhances the portability of detection operation so that the entire operation time is controlled within 1 h, and (iii) compared with common colorimetric-based RPA-ICA, the proposed assay used MQD to provide strong and quantifiable fluorescence signal, thus enhancing the detection sensitivity. With this strategy, the proposed RPA-MQD-based ICA can amplify and detect the SARS-CoV-2 nucleic acid on-site with a sensitivity of 2 copies/reaction, which is comparable to the sensitivity of commercial reverse transcription quantitative polymerase chain reaction (RT-qPCR) kits. Moreover, the designed primers did not cross-react with other common respiratory viruses, including adenovirus, influenza virus A, and influenza virus B, suggesting high specificity. Thus, the established portable method can sensitively detect SARS-CoV-2 nucleic acid without relying on equipment, having good application prospects in SARS-CoV-2 detection scenarios under non-lab conditions.


Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent of the coronavirus disease 2019 (COVID-19) pandemic [1][2][3][4], which has led to more than 605 million infections and 6.5 million deaths worldwide (https://covid19.who.int/; accessed on 5 September 2022). At present, direct virus detection methods are based on the sequence specificity of viral nucleic acid and the epitope specificity of viral capsid protein [5,6]. Due to the high sensitivity and specificity, nucleic acid-based reverse transcription quantitative polymerase chain reaction (RT-qPCR) is considered the gold standard approach [7,8]. RT-qPCR is based on the principle of complementary base pairing of nucleic acids and achieves exponential amplification of target fragments under the action of DNA polymerase. As the amplification reaction proceeds, the fluorescence signal gradually accumulates and can be read by the detector when the detection threshold is reached. Base pairing ensures the specificity, and exponential amplification and sensitive fluorescence signal ensure the sensitivity of the reaction. However, PCR-based methods still require a clean environment, large-scale instruments and professional technicians, thereby limiting their application in point-of-care testing (POCT) areas [9].

Nucleic Acid Release
In accordance with the instruction manual, viral nucleic acid templates were released by the RoomTemp Sample Lysis Kit for 10 min at room temperature. Virus lysates and dilutions were used in subsequent experiments.

Primer Design
We designed three pairs of RPA primers. Two pairs of primers were selected from the reference [34]. By verifying the above 5 pairs of primers one by one, only one pair of primers was screened to amplify SARS-CoV-2 without non-specific amplification. The amplified product (201 bp) is a part of the S protein gene. The optimal primers are as follows: 5 -Digoxin-CCACTGAGAAGTCTAACATAATAAGAGGCTG-3 , 5-Biotin-AATAAAC-TCTGAACTCACTTTCCATCCAACT-3

RT-RPA Assay
The reverse transcription RPA (RT-RPA) system contains pre-packaged and lyophilized enzyme mixture, 29.5 µL of Primer Free Rehydration buffer, 1 µL of each optimized primer (2.5 µM) for SARS-CoV-2, 2 µL of RNA template for SARS-CoV-2, 0.5 µL of ProtoScript II RT, 1 µL of Rnase H, 12.5 µL of double-distilled water and 2.5 µL of 280 mM magnesium acetate in a total volume of 50 µL. The nucleic acid amplification with a water bath at 37 • C was completed in 30 min.

Synthesis of SiO 2 -MQDs Nanocomposite
Firstly, the synthesis of SiO 2 NPs was carried out according to the method previously published by our laboratory [35]. Subsequently, the surface of SiO 2 NPs was labeled with quantum dots by PEI-mediated electrostatic adsorption. The preparation process was as follows: (1) to prepare PEI-coated SiO 2 NPs (SiO 2 /PEI), 1 mL of SiO 2 NPs (1 mg/mL) and 40 mL of aqueous PEI solution (0.2 mg/mL) were mixed and sonicated vigorously for one hour at room temperature; (2) after centrifugation (6000 rpm, 6 min), PEI-coated SiO 2 NPs were washed with deionized water three times; (3) the SiO 2 /PEI NPs were mixed and reacted with 40 mL of CdSe/ZnS-MPA QDs (1 nM) under ultrasonic conditions for 60 min at room temperature. Transmission electron microscopy (TEM) was used to evaluate whether quantum dots are absorbed on the surface of the SiO 2 NPs. After centrifugation (5400 rpm, 6 min) and washing, the prepared SiO 2 -QDs were resuspended in 1 mL of deionized water. The above steps were repeated twice to prepare SiO 2 -MQDs with high fluorescence intensity encapsulated by two-layer quantum dots (Dutta et al., 2022 [7]).

Preparation of Immuno-SiO 2 -MQDs Labels
Digoxin monoclonal antibody was conjugated with SiO 2 -MQDs via carbodiimide chemistry at room temperature [36]. First, 1 mL of SiO 2 -MQD was mixed with MES solution (0.1 M) with 1 mM EDC and 2 mM sulfo-NHS and then sonicated for 15 min. After centrifugation (4200 rpm, 6 min) and resuspension in 0.2 mL of PBS, the mixture was reacted with 15 µg of digoxin monoclonal antibody for two hours, followed by surface blocking with BSA for another one hour. Finally, immuno-SiO 2 -MQD labels were centrifuged (4200 rpm, 6 min), washed twice with PBS, redispersed in 1 mL of PBS solution, and stored at 4 • C.

Preparation of the ICA Strip
The ICA strip consisted of a sample pad, an NC membrane with one test line (T line), one control line (C line), and an absorbent pad. Streptavidin (1.0 mg/mL) and goat-antimouse IgG antibody (1.0 mg/mL) were spotted on the NC membrane to form the T line and the C line with a constant dispense of 0.1 µL/mm by the XYZ spraying platform (Biodot, Irvine, CA, USA) [37]. All the ICA components were placed into a drying oven at 37 • C for 2 h and then attached to the plastic backing card in sequence. Eventually, the assembled ICA card was cut into 3 mm-wide strips and preserved in a desiccator until use.

Analytical Procedure for Detection of SARS-CoV-2
An ICA strip was used to test the RPA products. First, RPA products (2 µL) were mixed with 2 µL immuno-SiO 2 -MQD labels and 70 µL running buffer solution (10 mM PBS, pH 7.4, 0.05% Tween-20, 1% BSA). Then, a sample pad of ICA strip was immersed into the mixture. Fifteen minutes later, the T line turned into a bright red line under the Handheld UV light (365 nm) illumination (Shenyu, Dongying, China) in the presence of the RPA products. Results could be read by the naked eye or photographed with a mobile phone.

Optimization of Experimental Procedures and Methodological Investigation
Considering that the reaction temperature of reverse transcription and RPA is similar, we tried to combine the two steps into one. In addition, the primer concentration and the amount of product detection were optimized. For methodological validation, we examined the specificity, sensitivity and reproducibility of the established method.

Construction of Immuno-MQD Labels and MQD-Based ICA
Herein, a high-performance MQD was prepared through the layer-by-layer assembly of highly photostable CdSe@ZnS-COOH QDs onto the surface of monodispersed SiO 2 NP, which can provide higher QD loads, better luminescence, good stability and dispersibility than traditional spherical fluorescent microspheres for ICA methods (Scheme 1 and Figure 1A-C). As observed in Figure 1D-G, this preparation method increases the fluorescence of the material. Moreover, the two-layer shell of SiO 2 -MQD increases the surface area of SiO 2 microspheres and provides more carboxyl groups, which is more conducive to the covalent attachment of antibodies [7]. As revealed in Figure 1E, the zeta potential of SiO 2 -based nanocomposites increased significantly after PEI combining and decreased sharply after QDs coating. The zeta potential values are 33.7, 36.6, 12.9, 38.9, and 12.3 mV for SiO 2 , SiO 2 -PEI, SiO 2 -QD, SiO 2 -QD-PEI and SiO 2 -MQD, respectively. Such a high and low potential change proves that the synthesis of MQD is driven by PEI-mediated electrostatic adsorption. When combined with antibody, the potential of immuno-SiO 2 -MQDs is significantly reduced ( Figure 1F).

Construction of Immuno-MQD Labels and MQD-Based ICA
Herein, a high-performance MQD was prepared through the layer-by-layer assembly of highly photostable CdSe@ZnS-COOH QDs onto the surface of monodispersed SiO2 NP, which can provide higher QD loads, better luminescence, good stability and dispersibility than traditional spherical fluorescent microspheres for ICA methods (Scheme 1 and Figure 1A-C). As observed in Figure 1D-G, this preparation method increases the fluorescence of the material. Moreover, the two-layer shell of SiO2-MQD increases the surface area of SiO2 microspheres and provides more carboxyl groups, which is more conducive to the covalent attachment of antibodies [7]. As revealed in Figure 1E, the zeta potential of SiO2-based nanocomposites increased significantly after PEI combining and decreased sharply after QDs coating. The zeta potential values are 33.7, 36.6, 12.9, 38.9, and 12.3 mV for SiO2, SiO2-PEI, SiO2-QD, SiO2-QD-PEI and SiO2-MQD, respectively. Such a high and low potential change proves that the synthesis of MQD is driven by PEI-mediated electrostatic adsorption. When combined with antibody, the potential of immuno-SiO2-MQDs is significantly reduced ( Figure 1F).

Construction of Immuno-MQD Labels and MQD-Based ICA
Herein, a high-performance MQD was prepared through the layer-by-layer assembly of highly photostable CdSe@ZnS-COOH QDs onto the surface of monodispersed SiO2 NP, which can provide higher QD loads, better luminescence, good stability and dispersibility than traditional spherical fluorescent microspheres for ICA methods (Scheme 1 and Figure 1A-C). As observed in Figure 1D-G, this preparation method increases the fluorescence of the material. Moreover, the two-layer shell of SiO2-MQD increases the surface area of SiO2 microspheres and provides more carboxyl groups, which is more conducive to the covalent attachment of antibodies [7]. As revealed in Figure 1E, the zeta potential of SiO2-based nanocomposites increased significantly after PEI combining and decreased sharply after QDs coating. The zeta potential values are 33.7, 36.6, 12.9, 38.9, and 12.3 mV for SiO2, SiO2-PEI, SiO2-QD, SiO2-QD-PEI and SiO2-MQD, respectively. Such a high and low potential change proves that the synthesis of MQD is driven by PEI-mediated electrostatic adsorption. When combined with antibody, the potential of immuno-SiO2-MQDs is significantly reduced ( Figure 1F).   The ICA for SARS-CoV-2 detection was designed and constructed by integrating four parts into a strip, including a sample pad, a test line coating the streptavidin, a control line coating goat anti-mouse IgG, and an absorbent pad (Scheme 2).
The ICA for SARS-CoV-2 detection was designed and constructed by integrating four parts into a strip, including a sample pad, a test line coating the streptavidin, a control line coating goat anti-mouse IgG, and an absorbent pad (Scheme 2). Scheme 2. Synthesis of SiO2-MQDs nanocomposite.

Optimization of RT-RPA and ICA
In order to shorten the experimental time and ensure accurate and readable experimental results, we have integrated the experimental steps and optimized the primer concentration and detection system.
RPA primers are longer than conventional PCR primers and prone to mismatch forming dimers or secondary structures, resulting in non-specific amplification. Therefore, before the template reaction, it is necessary to perform non-specific verification of different concentrations of the primers. As shown in Figure 2, primers with concentrations of 200 and 100 nM produced non-specific products, leading to false-positive results. When the primer concentration was reduced to 50 nM, a low-concentration RNA template (RNA-L, 80 copies) and high concentration RNA template (RNA-H, 800 copies) were added, respectively. It can be seen from the agarose gel electrophoresis that the size of the RPA product is consistent with the theoretical size (201 bp), and the ICA result was positive ( Figure 3A,B). Therefore, the forward and reverse primers concentrations were determined to be 50 nM for the SARS-CoV-2 assay.
Reverse transcription and RPA processes shorten the assay time. It is mentioned in the instructions for the reverse transcription kit and RPA kit that the reaction time of the two takes 60 and 30 min, respectively. We noticed that the temperature of the two reactions was about 40 °C , so we tried adding reverse transcriptase directly to the RPA system to synchronize reverse transcription and isothermal amplification. The experimental results in Figure 3A proved that the process was feasible and reliable.

Optimization of RT-RPA and ICA
In order to shorten the experimental time and ensure accurate and readable experimental results, we have integrated the experimental steps and optimized the primer concentration and detection system.
RPA primers are longer than conventional PCR primers and prone to mismatch forming dimers or secondary structures, resulting in non-specific amplification. Therefore, before the template reaction, it is necessary to perform non-specific verification of different concentrations of the primers. As shown in Figure 2, primers with concentrations of 200 and 100 nM produced non-specific products, leading to false-positive results. When the primer concentration was reduced to 50 nM, a low-concentration RNA template (RNA-L, 80 copies) and high concentration RNA template (RNA-H, 800 copies) were added, respectively. It can be seen from the agarose gel electrophoresis that the size of the RPA product is consistent with the theoretical size (201 bp), and the ICA result was positive ( Figure 3A,B). Therefore, the forward and reverse primers concentrations were determined to be 50 nM for the SARS-CoV-2 assay.
Reverse transcription and RPA processes shorten the assay time. It is mentioned in the instructions for the reverse transcription kit and RPA kit that the reaction time of the two takes 60 and 30 min, respectively. We noticed that the temperature of the two reactions was about 40 • C, so we tried adding reverse transcriptase directly to the RPA system to synchronize reverse transcription and isothermal amplification. The experimental results in Figure 3A proved that the process was feasible and reliable.
When optimizing the detection system, different amounts of MQD-labeled digoxin antibodies were used to compare the chromogenic effects. When the amount of template added was eight copies, and the primer concentration was 50 nM, the RPA products were used for testing. The added volume of labeled antibody in the sample loading system was 0.5, 1, 2, and 4 µL, respectively ( Figure 4). When the added volume was 2 and 4 µL, the T line can be clearly seen, but because the background color of the 4 µL system is high, the volume of antibody added was determined to be 2 µL. The ICA for SARS-CoV-2 detection was designed and constructed by integrating four parts into a strip, including a sample pad, a test line coating the streptavidin, a control line coating goat anti-mouse IgG, and an absorbent pad (Scheme 2). Scheme 2. Synthesis of SiO2-MQDs nanocomposite.

Optimization of RT-RPA and ICA
In order to shorten the experimental time and ensure accurate and readable experimental results, we have integrated the experimental steps and optimized the primer concentration and detection system.
RPA primers are longer than conventional PCR primers and prone to mismatch forming dimers or secondary structures, resulting in non-specific amplification. Therefore, before the template reaction, it is necessary to perform non-specific verification of different concentrations of the primers. As shown in Figure 2, primers with concentrations of 200 and 100 nM produced non-specific products, leading to false-positive results. When the primer concentration was reduced to 50 nM, a low-concentration RNA template (RNA-L, 80 copies) and high concentration RNA template (RNA-H, 800 copies) were added, respectively. It can be seen from the agarose gel electrophoresis that the size of the RPA product is consistent with the theoretical size (201 bp), and the ICA result was positive ( Figure 3A,B). Therefore, the forward and reverse primers concentrations were determined to be 50 nM for the SARS-CoV-2 assay.
Reverse transcription and RPA processes shorten the assay time. It is mentioned in the instructions for the reverse transcription kit and RPA kit that the reaction time of the two takes 60 and 30 min, respectively. We noticed that the temperature of the two reactions was about 40 °C , so we tried adding reverse transcriptase directly to the RPA system to synchronize reverse transcription and isothermal amplification. The experimental results in Figure 3A proved that the process was feasible and reliable.    When optimizing the detection system, different amounts of MQD-labeled digoxin antibodies were used to compare the chromogenic effects. When the amount of template added was eight copies, and the primer concentration was 50 nM, the RPA products were used for testing. The added volume of labeled antibody in the sample loading system was 0.5, 1, 2, and 4 μL, respectively (Figure 4). When the added volume was 2 and 4 μL, the T line can be clearly seen, but because the background color of the 4 μL system is high, the volume of antibody added was determined to be 2 μL. When the amount of template added was 8 copies and the primer concentration was 50 nM, the RPA products were used for testing. (B) When the added volume was 2 μL, the fluorescence intensity difference between the background and control was the most significant. Error bars represent the standard deviation of three repetitive experiments.

Specificity of the RT-RPA-Combined ICA
Since the clinical symptoms caused by SARS-CoV-2 and other respiratory viruses are similar [34], cross-reactivity with other viruses needs to be examined. Inactivated SARS-CoV-2, IAV, IBV, and ADV were adjusted to the same concentration of 10 3 copies/μL and were amplified in accordance with the above process to verify the assay specificity. Each sample was independently repeated thrice, and the image was recorded and processed to calculate its signal value. As seen in Figure 5, the RPA product of the SARS-CoV-2 shows a bright T line, while other respiratory viruses do not show positive results. The result shows that the designed system has excellent specificity for SARS-CoV-2.  When optimizing the detection system, different amounts of MQD-labeled digoxin antibodies were used to compare the chromogenic effects. When the amount of template added was eight copies, and the primer concentration was 50 nM, the RPA products were used for testing. The added volume of labeled antibody in the sample loading system was 0.5, 1, 2, and 4 μL, respectively (Figure 4). When the added volume was 2 and 4 μL, the T line can be clearly seen, but because the background color of the 4 μL system is high, the volume of antibody added was determined to be 2 μL. When the amount of template added was 8 copies and the primer concentration was 50 nM, the RPA products were used for testing. (B) When the added volume was 2 μL, the fluorescence intensity difference between the background and control was the most significant. Error bars represent the standard deviation of three repetitive experiments.

Specificity of the RT-RPA-Combined ICA
Since the clinical symptoms caused by SARS-CoV-2 and other respiratory viruses are similar [34], cross-reactivity with other viruses needs to be examined. Inactivated SARS-CoV-2, IAV, IBV, and ADV were adjusted to the same concentration of 10 3 copies/μL and were amplified in accordance with the above process to verify the assay specificity. Each sample was independently repeated thrice, and the image was recorded and processed to calculate its signal value. As seen in Figure 5, the RPA product of the SARS-CoV-2 shows a bright T line, while other respiratory viruses do not show positive results. The result shows that the designed system has excellent specificity for SARS-CoV-2. When the amount of template added was 8 copies and the primer concentration was 50 nM, the RPA products were used for testing. (B) When the added volume was 2 µL, the fluorescence intensity difference between the background and control was the most significant. Error bars represent the standard deviation of three repetitive experiments.

Specificity of the RT-RPA-Combined ICA
Since the clinical symptoms caused by SARS-CoV-2 and other respiratory viruses are similar [34], cross-reactivity with other viruses needs to be examined. Inactivated SARS-CoV-2, IAV, IBV, and ADV were adjusted to the same concentration of 10 3 copies/µL and were amplified in accordance with the above process to verify the assay specificity. Each sample was independently repeated thrice, and the image was recorded and processed to calculate its signal value. As seen in Figure 5, the RPA product of the SARS-CoV-2 shows a bright T line, while other respiratory viruses do not show positive results. The result shows that the designed system has excellent specificity for SARS-CoV-2.

Sensitivity of the RT-RPA-Combined ICA
The sensitivity test shows the minimum detection limit of the method. After viral lysis, the released RNA templates of SARS-CoV-2 were accurately quantified by digital PCR (QIAGEN, Redwood City, CA, USA) with the concentration of 4 × 10 6 copies/µL (see Supplementary Material S1). Then, RNA templates to be subjected to RT-RPA assay were serially diluted to simulate different concentrations of viral nucleic acid samples. As shown in Figure 6A, the red T line becomes weaker as the template concentration decreases. A fourparameter curve is fitted based on signal intensity and template concentration in Figure 6B. The IUPAC protocol (LOD = yblank + 3SDblank) was adopted to determine the limits of detection (LOD) of this assay, where yblank and SDblank are the average fluorescence intensity and standard deviation of the blanks, respectively [35,36,38]. Therefore, the LOD (before RPA amplification but not before immunodetection) for SARS-CoV-2 was estimated as two copies per reaction in this assay. The RT-qPCR kit (Daan Gene, Guangzhou, China) was used to detect the same SARS-CoV-2 RNA templates with an LOD of 20 copies for the N gene and 2 copies for the ORF1a/b gene per reaction in Figure 6C [37]. It can be seen that the detection scheme established in this experiment can achieve the same sensitivity as the commercial SARS-CoV-2 RT-qPCR kits. However, the scheme established in this study is fast, portable, and does not need to rely on instruments. It may have very important applications in limited scenarios, such as primary hospitals, underdeveloped areas, and self-examination at home.

Sensitivity of the RT-RPA-Combined ICA
The sensitivity test shows the minimum detection limit of the method. After viral lysis, the released RNA templates of SARS-CoV-2 were accurately quantified by digital PCR (QIAGEN, Redwood City, CA, USA) with the concentration of 4 × 10 6 copies/μL (see Support Material S1). Then, RNA templates to be subjected to RT-RPA assay were serially diluted to simulate different concentrations of viral nucleic acid samples. As shown in Figure 6A, the red T line becomes weaker as the template concentration decreases. A fourparameter curve is fitted based on signal intensity and template concentration in Figure  6B. The IUPAC protocol (LOD = yblank + 3SDblank) was adopted to determine the limits of detection (LOD) of this assay, where yblank and SDblank are the average fluorescence intensity and standard deviation of the blanks, respectively [35,36,38]. Therefore, the LOD (before RPA amplification but not before immunodetection) for SARS-CoV-2 was estimated as two copies per reaction in this assay. The RT-qPCR kit (Daan Gene, Guangzhou, China) was used to detect the same SARS-CoV-2 RNA templates with an LOD of 20 copies for the N gene and 2 copies for the ORF1a/b gene per reaction in Figure 6C [37]. It can be seen that the detection scheme established in this experiment can achieve the same sensitivity as the commercial SARS-CoV-2 RT-qPCR kits. However, the scheme established in this study is fast, portable, and does not need to rely on instruments. It may have very important applications in limited scenarios, such as primary hospitals, underdeveloped areas, and self-examination at home.

Repeatability of the RT-RPA-Combined ICA
In order to examine the repeatability of this protocol, 40 copies of the SARS-CoV-2 RNA template were used for detection. The detection was performed 10 times, and the detection rate was 100% (Figure 7). The fluorescence signal values of 10 tests were measured and counted, the average signal value was 25.2, and the relative standard deviation (RSD%) was 19%.

Repeatability of the RT-RPA-Combined ICA
In order to examine the repeatability of this protocol, 40 copies of the SARS-CoV-2 RNA template were used for detection. The detection was performed 10 times, and the detection rate was 100% (Figure 7). The fluorescence signal values of 10 tests were measured and counted, the average signal value was 25.2, and the relative standard deviation (RSD%) was 19%.

Conclusion
In this study, we have combined RPA nucleic acid amplification technology and multilayer quantum dot-based immunochromatography technology to develop a fast and convenient on-site detection solution for SARS-CoV-2. The scheme does not rely on instruments and can amplify viral nucleic acid with the body temperature (37 °C ) of the operator. Compared with traditional nucleic acid extraction and detection technology, which takes at least 3 h, the whole process of this scheme can be completed within 1 h, with similar sensitivity. In addition, compared with RT-qPCR, the cost of RT-RPA is mainly the RPA kit (less than USD3 per reaction), and the instrument cost can be ignored. We believe that the scheme established in this study may provide a reference for rapid, portable, low-cost and sensitive detection of SARS-CoV-2 viruses.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: Raw results of Digital PCR; Figure S2: One-dimensional distribution map of each sample well of F1 channel (ORF1a/b gene); Figure S3: One-dimensional distribution map of each sample well of N channel (N gene); Figure S4: Two-dimensional distribution map of F1 channel (ORF1a/b) and N channel (N gene).

Conclusions
In this study, we have combined RPA nucleic acid amplification technology and multilayer quantum dot-based immunochromatography technology to develop a fast and convenient on-site detection solution for SARS-CoV-2. The scheme does not rely on instruments and can amplify viral nucleic acid with the body temperature (37 • C) of the operator. Compared with traditional nucleic acid extraction and detection technology, which takes at least 3 h, the whole process of this scheme can be completed within 1 h, with similar sensitivity. In addition, compared with RT-qPCR, the cost of RT-RPA is mainly the RPA kit (less than USD3 per reaction), and the instrument cost can be ignored. We believe that the scheme established in this study may provide a reference for rapid, portable, low-cost and sensitive detection of SARS-CoV-2 viruses.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/pathogens11111252/s1, Figure S1: Raw results of Digital PCR; Figure S2: One-dimensional distribution map of each sample well of F1 channel (ORF1a/b gene); Figure S3: One-dimensional distribution map of each sample well of N channel (N gene); Figure S4: Two-dimensional distribution map of F1 channel (ORF1a/b) and N channel (N gene).  Data Availability Statement: The datasets generated during and/or analyzed during the current study can be find in the main text and the supplementary materials.