An updated review of SARS‐CoV‐2 detection methods in the context of a novel coronavirus pandemic

Abstract The World Health Organization has reported approximately 430 million confirmed cases of coronavirus disease 2019 (COVID‐19), caused by severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2), worldwide, including nearly 6 million deaths, since its initial appearance in China in 2019. While the number of diagnosed cases continues to increase, the need for technologies that can accurately and rapidly detect SARS‐CoV‐2 virus infection at early phases continues to grow, and the Federal Drug Administration (FDA) has licensed emergency use authorizations (EUAs) for virtually hundreds of diagnostic tests based on nucleic acid molecules and antigen–antibody serology assays. Among them, the quantitative real‐time reverse transcription PCR (qRT‐PCR) assay is considered the gold standard for early phase virus detection. Unfortunately, qRT‐PCR still suffers from disadvantages such as the complex test process and the occurrence of false negatives; therefore, new nucleic acid detection devices and serological testing technologies are being developed. However, because of the emergence of strongly infectious mutants of the new coronavirus, such as Alpha (B.1.1.7), Delta (B.1.617.2), and Omicron (B.1.1.529), the need for the specific detection of mutant strains is also increasing. Therefore, this article reviews nucleic acid‐ and antigen–antibody‐based serological assays, and compares the performance of some of the most recent FDA‐approved and literature‐reported assays and associated kits for the specific testing of new coronavirus variants.

stranded RNA genome (26-32 kb). 3 There are four genera of coronaviruses, α, β, γ, and δ, and both SARS-CoV-2 and SARS-CoV are members of the beta coronavirus family, while Middle East respiratory syndrome coronavirus (MERS-CoV) belongs to family C of the genus β coronavirus. SARS-CoV-2 and SARS-CoV share 79.6% sequence similarity, and research has revealed that these two viruses share the same vascular angiotensin-converting enzyme 2 (ACE2) receptor for infection of human cells. 4 SARS-CoV-2 is circulated primarily through breathing or contact with droplets from an infected person, with a latency period of about 2-14 days. 5 The patient's clinical presentation after infection varies from asymptomatic to severe, with most infections not being severe. 6 The leading causes of death commonly associated with COVID-19 are respiratory failure, followed by septic shock, renal failure, hemorrhage, and cardiac failure. 7 Thousands of cumulative mutations of the SARS-CoV-2 have occurred since its emergence, which often occurs naturally during rep- designing primers that target their mutation sites. In addition, with the development of identifiable conserved protein tag tails, the detection rate of POC-based immunoassay assays is also increasing. The development of POC assays is expected to be applied in the future in communities, rural areas, and other relatively poorly resourced areas for effective epidemic control. 11 Moreover, the optimization of samples and swabs and other sampling tools, as well as the combination of artificial intelligence and deep learning networks, are also worth considering in the development of POC assays.
Herein, we comprehensively review the practical techniques designed to detect SARS-CoV-2, evaluate the results of relevant technologies (Table 1), and enumerate the relevant FDA-approved test kits and the latest mutant detection devices.
2 | STRUCTURE AND DETECTION OF SARS-COV-2 2.1 | The structure and biology of SARS-CoV-2 Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is in the genus beta coronavirus and is the seventh coronavirus to infect humankind and cause acute respiratory disease. 12 SARS-CoV-2 is 60-140 nm in diameter and comprises a single-stranded positive-sense RNA genome, capsid protein, and outer membrane assembly. Its genome size is from 29.8 to 29.9 kb and it includes 14 open reading frames (ORFs), which encode 27 proteins. 13 Its genome is almost 80% homologous to SARS-CoV and is similar to bat coronavirus (bat CoV), with 96% sequence similarity. Among the ORFs, ORF1ab, located in the 5 0 -untranslated region (UTR), is the largest gene, encoding a variety of proteins required for viral transcription and replication, including multiple nonstructural proteins (NSP). The gene located in the 3 0 -UTR encodes four predominant structural proteins, including the spike (S) protein, the membrane (M) protein, the envelope (E) protein, and the nucleocapsid (N) protein, and also encodes many nonstructural proteins. 14 Among the four structural proteins, the S protein serves as a transmembrane protein that can mediate coronavirus entrance into the host cell by interacting with angiotensin-converting enzyme 2 (ACE2). 15 The M protein plays a role in determining the configuration of the viral envelope and the assembly of viral particles, and also counteracts the innate antiviral immune response triggered by viral RNA. 16 The N proteins can combine with the RNA genome of viruses to constitute N protein-RNA complexes that participate in the replication cycle of the virus, the host response to viral infection, and genomic signaling. Meanwhile, the E protein, as the minimal major structural protein, can interact with host cell membrane proteins to participate in the viral production and the maturation process. 17

| Infection and sample collection of SARS-CoV-2
The S protein is a trimeric class I viral fusion protein that has a critical function in mediating the adhesion, fusion, and entry of SARS-CoV-2 into the human body. [18][19][20] S protein has two subunits, S1 and S2. The S1 subunit can bind to the ACE2 receptor on the host cell and contains both an N-terminal domain (NTD) and a receptor-binding domain (RBD). The RBD of the S1 subunit carries out a hinge-like motion F I G U R E 1 Biology and serology of SARS-CoV-2 infection (a) Structure and infection: SARS-CoV-2 is an RNA virus that consists of four structural proteins, the Spike (S) protein, Nucleocapsid (N) protein, Membrane (M) protein, and Envelope (e), together with many nonstructural proteins to maintain the biological traits of the virus.
Step 1-3: S protein allows the virus to bind and enter human cells and consists of S1 and S2 subunits. S1 can bind the angiotensin-converting enzyme 2 (ACE2) receptor. After S1 binds to ACE2, S protein is hydrolyzed by the action of TMPRSS2 protease.  15,21 The S1/S2 protease cleavage site exists between the S1 and S2 subunits, and the host protease can cleave the S protein at the S2 0 site, which activates the protein and fuses the virus to the host cell membrane through irreversible conformational changes (Figure 1a). 22 SARS-CoV-2 is extraordinarily stable at 4 C for 14 days and can be viable at 37 C for 24 h. 23 It can be transmitted by respiratory secretions, aerosols, direct contact, the fecal-oral route, mother-tochild transmission, and ocular transmission. 24,25 Infected individuals usually begin to show symptoms within 8.2-15.6 days, with an average of 11.2 days, with the disease progressing more rapidly in the elderly than in younger people. 26 After human infection, the virus deposits in the upper respiratory tract and gradually penetrates deep into the lungs; however, the virus can also cause damage to the nervous system (e.g., the brain), digestive system (e.g., the liver, stomach, intestines), the urinary system (e.g., the kidneys), and the cardiovascular system. 27 Viruses can provoke an immune reaction in the body, with immunoglobulin M (IgM) as the first line of protection, usually appearing within 3-5 days after infection. Immunoglobulin G (IgG) often appears 1 week after infection, with high affinity and adaptive response, and a long duration, making it useful as a marker of the previous infection ( Figure 1c). There are two principal categories of SARS-CoV-2 tests adopted currently: (1) Nucleic acid-based viral tests; and (2) antigen-and antibody-based serological viral tests.  33 In the qRT-PCR protocol, reverse transcriptase converts the extracted and purified SARS-CoV-2 RNA into cDNA, which is then amplified using gene-specific primers in the quantitative real-time PCR step of the qRT-PCR protocol. Repeated thermal cycling in which the probe reports a fluorescent signal each time the target region of the genome is amplified results in quantitative detection ( Figure 2a). 34 Viral RNA extraction is now commonly performed by making use of upper airway specimens (e.g., nasopharyngeal swabs or oropharyngeal swabs, which were used more frequently) and lower airway specimens (e.g., phlegm and bronchoalveolar lavage fluid), but also blood, stool, and tissue samples. qRT-PCR can target regions such as ORF1ab (RdRp), N, E, S, and ORF8 genes, among which the RdR1ab located in RdRP, and the N and E gene in OFR1ab are more conserved, with the detection of the RdRP and E genes being less restrictive and more sensitive compared with N gene detection. 35 The WHO developed and shared primers that target the E gene, as well as the RdRp gene sequence, to screen for and confirm SARS-CoV-2 for the first time worldwide, and the design method based on this also successfully differentiated SARS-CoV and SARS-CoV-2. 36   used a Bayesian hierarchical model to analyze 1330 confirmed cases to assess the false-negative rate between 5 days before symptom onset and 21 days after the occurrence of symptoms, and found that the false-negative rate on the day before the symptoms appear, the day symptoms appear, and the 21st day after symptom onset were 67%, 38%, and 66%, and the median false-negative rate gradually decreased to 20% on days 3 and 4 of symptom onset. 40 assay kits were evaluated. The overall mean Ct value of the five kits F I G U R E 2 Nucleic acid-based detection of SARS-CoV-2 (a) qRT-PCR: Step 1-4: SARS-CoV-2 RNA in different collected samples, such as nasopharyngeal swabs, can be extracted and purified using an RNA extraction kit, and complementary DNA (cDNA) for amplification and detection can be obtained by reverse transcriptase; Step 5-9: template cDNA undergoes denaturation, primer annealing, and extension in the real-time PCR instrument The fluorescence signal is released when the fluorescence molecule is no longer inhibited by the quenching molecule, and the instrument can convert the fluorescence signal in the cycle into the cycle threshold (CT) value, which can be expressed as the quantified viral load data, and the validity of SARS-CoV-2 infection is verified by comparison with negative controls and threshold lines. (b) CRISPR/Cas system: Based on reverse transcription recombinant polymerase amplification (RT-RPA) and reverse transcription loop-mediated isothermal amplification (RT-LAMP), purified RNA can be amplified in an isothermal instrument, and the amplified product can be reported both by the chromogenic substances in the amplification system and by the CRISPR/Cas system for further specific cleavage of nucleic acids and determination of virus infection. The CRISPR-associated Cas protein then binds to the guide RNA, forming a complex that can target cleavage of the viral nucleic acid sequence, and the result can be reported by the fluorescence quenching molecules in the reaction, by reporting the fluorescence signal, or by the side stream chromatography color development strip of the cleaved nucleic acid fragment. was 23.6 ± 3.8, with accuracy ranging from 96.9% to 100%, among which the SARS-CoV-2 Variants II Assay Allplex (for L452R, W152C, K417T, K417N) kits had 100% sensitivity and specificity. 44

| PCR technique used in variants detection
Outbreaks are difficult to control because of the high infectivity of mutant strains, and the emergence of mutant strains can adversely affect the performance of molecular assays, especially those targeting genomic single-target tests. N and E genes as targets, while the S gene is often off-target due to its susceptibility to mutation. In one study, a mutation in the viral genome at locus 26,340, C to U, caused a failure of the cobas SARS-CoV-2 E gene qRT-PCR assay, but because the detection probe of the cobas SARS-CoV-2 qRT-PCR kit can target both regions of the genome, the experimenter was still tested positive, which also reminded researchers to develop Multiple-target primer sets were developed to avoid falsenegative results. 45 54 Primerprobe targets against SARS-CoV-2 ORF1ab and S genes have also been reported. 54 The ORF1b region was also selected for LAMP amplification using six primers and the results obtained were verified by gel electrophoresis.

| Evaluation of RT-LAMP detection results
LAMP-based assays are available in tiny PCR tubes, where dumbbelllike structures with many DNA synthesis initiation sites can be transferred into longer tandems (where each tandem has many DNA synthesis initiation sites) during nucleic acid amplification, eventually leading to the accumulation of different DNA structures with the same target DNA sequence, 33 which in turn can be determined by turbidity, the addition of pH-sensitive dyes, or intercalation dyes to produce color or fluorescence; agarose gel electrophoresis of the products can also be used to determine SARS-CoV-2 infection. 55 RT-LAMP uses more primers than RT-PCR; therefore, it has a higher specificity. 51 The LAMP procedure is up to 10 times more sensitive than routine PCR for the assay of new coronaviruses in the absence of false negatives. Yu et al. also designed a LAMP-based diagnosis technique for SARS-CoV-2 testing using six primers, termed iLACO (isothermal LAMP-based method for COVID- 19), and found that the sensitivity and accuracy of iLACO were better than that of the Taqman-based qPCR detection method. 56 Recently, it was also shown that RT-LAMP targeting the SARS-CoV-2 N gene could specifically detect viral RNA of SARS-CoV-2 without cross-reactivity with related coronaviruses. (e.g., MERS-CoV, HCoV-229E) and other viruses that can lead to respiratory illnesses (e.g., RSVA, RSVB, and ADV). 57 These results also suggest that RT-LAMP-based technology has a promising prospect in the diagnosis of SARS-CoV-2 infection.  cron. 70 The POC-based miSHERLOCK CRISPR/Cas suite for the S protein mutation sites N501Y, Y144del, and E484K was also demonstrated to detect Alpha, Beta, and Gamma variants. 71

| Analysis of nucleic acid-based SARS-CoV-2 detection and other methods
Although real-time RT-qPCR is considered the gold standard method and the most widely applied in most countries, the detection protocols of all mentioned above need expensive experimental instruments, reagents, professional laboratories, and researchers. What's more, the accuracy of test results depends a lot on sample types 72 Figure 3).
The human body produces specific antibodies against SARS-CoV-2 infection, and these antibodies can be used as targets for the fast, simple, and highly sensitive detection of the virus with a sensitivity of >57.2% and up to 87.5% for IgM and >71.4% and up to 87.5% for IgG. 80 Notably, the RBD of the S protein displays higher antigenicity than the N protein, as shown by studies showing sensitivities of 96.8%, 96.8%, and 98.6% for RBD IgM, IgG, and IgA, respectively. 81 Many experts recommend the detection of specific antibodies as a supplement to nucleic acid testing, and paper-based lateral flow immunoassays (LFIA) have been developed (Figure 3a). antigen rapid test Cassette, the SureScreen-V kit, the Encode kit, and the E25Bio rapid diagnostic test. The specificity, LoD, and sensitivity F I G U R E 3 Serological detection of SARS-CoV-2 (a) Lateral flow assay: Quantum dots/colloidal gold can couple antibodies via specific labeling (using agent Maleamide-polyethylene glycol-succinimide ester (SMPEG)) and nonspecific labeling (using EDC/NHS chemistry methods). The rapid quantum dot and colloidal gold immunodiagnostic method for SARS-CoV-2 antibody-based on high specificity recombinant protein and quantum dot/colloidal gold immunofluorescence probes by double antibody sandwich or indirect method methodology using lateral flow assay. The patient sample added to the sample pad will move to the absorbent pad along the NC membrane by chromatography, which will form the tagged-antibody-antigen-antibody complex. After 10-15 min, test results can be observed on the test kit and operators can get an accurate fluorescence signal by a handheld fluorescent immunoanalyzer. (b) Cloud Network Platform: Rapid test kits can be used at the point of care for suspicious population screening tests, mobile devices such as cell phones can be used for result identification, handheld fluorescent immunoassay analyzers can perform a quantitative and qualitative analysis of test results, and qualitative and quantitative data can be uploaded to the terminal database, the CDC can manage relevant infections and suspicious populations through analysis of qualitative and quantitative data, give relevant clinical diagnosis recommendations, and combine with wearable devices such as smartwatches to achieve daily monitoring of people's medication, body temperature, heart rate, and other vital signs at the point of care such as communities and families, to control the development of epidemics in a timely and effective manner.

| Enzyme-linked immunosorbent assay
ELISA is considered the gold standard for laboratory testing for SARS-CoV-2. Using serological samples, the S protein (consisting of the S1 and S2 subunits, and the RBD) and the N protein of the virus can be used as the major immunogens to assay for serum virus-neutralizing antibodies in patients, 91,92 which can assay immunoglobulins of the virus in samples. 93 ELISA for virus detection is based on the antigenantibody complex structure and enzyme-labeled antibodies, among which indirect ELISA and sandwich ELISA are the two most commonly used methods of detection. 94 The enzyme on the enzyme-labeled antibody can catalyze the hydrolysis, oxidation, and reduction of the substrate to form a colored substance, which can be analyzed qualitatively by the naked eye or quantitatively by a spectrometer or other device, 95 where the strength of the colored signal is proportional to the level of the antigen or antibody is detected.
The patient's antibody levels, as well as the SARS-CoV-2 protein as an antigen, are two important factors affecting serological testing.

Most patients infected with the new coronavirus develop specific
IgM, IgA, and IgG responses within days 5-15, with IgM and IgA lasting 3-6 weeks and IgG lasting several months. 96,97 Recently, an ELISA kit was developed using the RBD region from S protein, which had a specificity of 99.3% and could detect a large number of antibodies 2 weeks after the appearance of symptoms. 98 ELISAs to assay IgG and IgM antibodies using the N and S proteins of the new coronavirus have been developed and the positive detection rates for the S protein-based ELISA and the N protein-based ELISA were 82.2% and 80.4%, with the S protein-based ELISA being significantly more sensitive to IgM than the N protein-based ELISA. 92

| Mutation sites on mutant strains cause antibody capture evasion in serological assays
The N protein is highly immunogenic and is the most produced pro- These also included A376T coupled to M241I and the most common A220V mutation, which escaped detection by capture antibodies and gave false-negative Abbott PanbioTM COVID-19 Ag assay results. 100 Given that the N antigen or "S antigen + N antigen" is mostly used as a marker in current serological kits, we point out that mutated sites in mutant strains may escape antibody capture, leading to reduced sensitivity and false-negative results. In Omicron, for example, there are 32 mutant sites on the S protein, including N501Y, L452, K477, and E484, which have been shown to evade serum-neutralizing antibody binding. [101][102][103] For the "S antigen + N antigen" assay kit, the presence of a large number of mutations on the S protein can cause a significant decrease in assay sensitivity and lead to false-negative results in serological assays. Therefore, we suggest that researchers evaluate and validate currently available antigen detection kits using VOCs samples and develop neutralizing antibodies based on conserved epitopes to improve the sensitivity of antigen detection kits.

| Analysis of antigen-antibody-based serological SARS-CoV-2 detection
In general, antigen-antibody-based serological SARS- were taken into consideration. 108 Therefore, WHO suggests that Ag-RDTs tend to conditions that are remote and underserved or seriously pandemic.
Given the average time of immune response to SARS-CoV-2 is around 1-2 weeks, the span of immune response will influence the clinical diagnosis. In the post-pandemic era, vaccination will gradually cover most people, which will complicate the results of antibody detection. Therefore, the applicable conditions of antibody detection should be considered (Table 3).

| MULTI-CHANNEL DETECTION OF SARS-COV-2 AND OTHER RESPIRATORY INFECTIOUS DISEASES
In patients. 124 Not only can the fluorescence sensor detect the strip in 10 min, but also connect to edge hardware devices (personal computers, smartphones, IPTV, etc.) and the fog layer of the network to perform reliable data transmission with low latency and high security.
What's more, several COVID-19 monitoring mHealth applications were proposed, which enabled patients to record and upload their results. 125 In the online hyper-connected world, the SARS-CoV-2 epidemic can be predicted through the sharing and analysis of medical data, including mathematical prediction models and algorithms (Figure 3b).
In conclusion, with the use of fast and accurate POC biosensing equipment, the detection results are uploaded to the mobile cloud monitoring platform in real time, which in turn establishes a cloudbased big data quality management and epidemic spread control system, generating a dynamic map of virus epidemic development control from two dimensions, spatial and temporal, so that the CDC command center can fully and timely understand the instantaneous information changes of the epidemic prevention and control grassroots units to achieve efficient and rapid linkage and unified Coordinated scheduling and resource allocation, thus effectively controlling the epidemic.

| CONCLUSION AND PROSPECT
At present, a range of nucleic acid molecule and antigen-antibody based methods are accessible for SARS-CoV-2 detection. The highly specific and sensitive nature of nucleic acid testing has led to its use in many countries for high-throughput analysis of numerous specimens in the population; but because of its equipment, space, and personnel requirements, nucleic acid testing can only be performed in specialized sites such as hospitals and CDCs. Serology-based test kits can meet the need for home and community-based POC testing because of their small size, flexibility, and less demanding testing environment. The recently emerged antigen-antibody test kits with high sensitivity and specificity can also serve as a supplement to detect SARS-CoV-2 outbreaks caused by strong mutant strains, such as Delta and Omicron, as well as for outbreak control in home and community care settings.
qRT-PCR continues to be the mainstream gold standard method to detect SARS-CoV-2 qualitatively and quantitatively. Nevertheless, the assay still has limitations, such as differences in viral load in various samples that affect the sensitivity of the assay, and mutation sites generated in mutant strains that affect the binding of primers and detection antibodies in serological kits. Highly infectious SARS-CoV-2 mutants and asymptomatic patients with false-negative test results also present a requirement for fast, highly sensitive, highly specific, and cost-effective POC-based testing kits. LAMP-and CRISP/Cas-