Unlike traditional food testing methods, the detection of 与传统食品检测方法不同，冷链食品上SARS-CoV-2 on cold-chain foods puts forward higher requirements regarding sampling method, detecting conditions, testing periodic time, anti-interference capability, and portability of equipment ^[1]. Here, according to the different pathogen markers, e.g., nucleic acid, antigen, antibody, and cytokine storm assay, researchers list two categories of 的检测对采样方法、检测条件、检测周期时间、抗干扰能力、设备便携性提出了更高的要求[48]。在这里，根据不同的病原体标志物，例如核酸，抗原，抗体和细胞因子风暴测定，我们列出了两类COVID-19 testing methods, as well as the advance of 检测方法，以及SARS-CoV-2 test methods (Table 检测方法的进步（表1). Whether the existing detection methods can be applied to test suspicious cold-chain foods will be dialectically discussed.）。现有的检测方法是否可以应用于检测可疑的冷链食品，将进行辩证讨论。

在核酸检测方法中，RT-PCR can effectively amplify trace viral genes in nucleic acid mixtures and has the characteristics of fast detection speed, high sensitivity, and strong specificity ^[37]. 能有效扩增核酸混合物中的痕量病毒基因，具有检测速度快、灵敏度高、特异性强等特点[82]。RT-PCR uses sequence-specific primers to identify tiny 使用序列特异性引物来鉴定微小的RNA targets. The recognized RNA is then transcribed by reverse transcriptase to cDNA, which is then used as a template for DNA靶标。然后通过逆转录酶将识别的RNA转录为cDNA，然后将其用作通过PCR进行DNA复制的模板。然而，测量是一种酶依赖性多步技术，操作复杂。周转时间需要几个小时，无法满足冷链食品快速检测的要求[83]。检测设施和仪器不便于携带，采集到的样本需要运到实验室进行检测。由于收集或处理不当，样品可能会产生假阴性结果。操作不当或实验室条件不足可能由于气溶胶污染而导致假阳性[84，85，86]。一步式嵌套 replication through PCR. However, the measurement is an enzyme-dependent multi-step technique, and the operation is complicated. The turnaround time takes a few hours which cannot meet the requirements of rapid testing of cold-chain food ^[38]. The test facilities and instruments are not portable, and the collected samples need to be transported to the laboratory for testing. The samples may produce false-negative results due to improper collection or processing. Improper operation or insufficient laboratory conditions may cause false positives due to aerosol contamination ^[39][40][41]. One-step nested RT-T-PCR is是一种灵活简便的 a flexible and easy method to test SARS-CoV-2. If the coronavirus mutates in one key amplified nucleotide, at least one pair can still be amplified ^[42]. The detection cost is lower than 检测方法。如果冠状病毒在一个关键的扩增核苷酸中发生突变，则至少一对核苷酸仍可被扩增[87]。检测成本低于RT-PCR, but nested ，但嵌套PCR is not feasible as a detection method for cold-chain food, because it is time-consuming and has a high risk of cross-contamination. In contrast, repetitive digital PCR is less interfered with by background wild DNA molecules, so it can reduce the impact of non-target DNA during cold-chain detection ^[43]. Digital 作为冷链食品的检测方法不可行，因为它耗时且具有较高的交叉污染风险。相比之下，重复性数字PCR受背景野生DNA分子的干扰较小，因此可以减少冷链检测过程中非靶DNA的影响[88]。当冷链食品样品的病毒载量较低或样品核酸降解时，数字PCR has obvious advantages when the viral load of cold-chain food samples is low or the sample nucleic acid is degraded. However, the cost of digital 具有明显的优势。然而，数字PCR detection is relatively high and the instrument is not portable ^[44]. It also involves the usage and storage of enzyme reagents, which will increase the cost and technological hurdles of detection. The digital 检测的成本相对较高，且仪器不便于携带[89]。它还涉及酶试剂的使用和储存，这将增加检测的成本和技术障碍。数字PCR operation process has the disadvantage of being easily contaminated. To avoid false-positive results, it is necessary to establish strict internal quality control specifications for the laboratory and strictly regulate the testing operation process ^[45][46]. 操作过程具有容易被污染的缺点。为避免假阳性结果，有必要为实验室建立严格的内部质量控制规范，并严格规范检测操作流程[90，91]。

Reverse transcription loop-mediated isothermal amplification (RT-LAMP) is a nucleic acid amplification assay, which is characterized by multiple specific primers for the target gene at a constant temperature of 60–65 °C under the DNA polymerase. About 10⁹-10¹⁰ times of nucleic acid amplification can be achieved in 15–60 min. RT-LAMP has the characteristics of simple operation, strong specificity (2 to 5 orders of magnitude higher than traditional PCR methods), and easy product detection ^[47][92]. The RT-LAMP results can be judged by visually observing the generation of white turbidity or green fluorescence. It is simple, quick, and does not require gel electrophoresis like PCR. As a point-of-care testing (POCT)-type nucleic acid detection method, LAMP requires less professional equipment, such as a thermal cycler, and the price of the instrument is lower than qRT-PCR, which can well meet the real-time requirements of SARS-CoV-2 detection on cold-chain food ^[48][93]. The detection sensitivity of RT-LAMP can reach 10 copies, and it has high specificity ^[49][94]. RT-LAMP is highly suitable for detecting > 60 copies/10 μL in sample. However, the test involves the use of enzymes (e.g., recombinase) that the storage of enzyme reagent needs, resulting in extra cost. When the test is performed for a long time, non-specific amplification may produce false-positive results if cold-chain food sampling < 10 copies/10 μL ^[50][95]. Based on RT-LAMP, a portable and scalable laser-engraved microwell array chip for multiplex amplification of viral RNA samples has been developed, which is a promising device for SARS-CoV-2 detection on cold-chain food.

CRISPR-Cas is a nonspecific RNA system that can be activated by the amplified product RNA, cleavages the reporter RNA, and releases a fluorescent dye from the quencher. The CRISPR-based system has attracted growing enthusiasm due to its pathogen diagnosis ability ^[51][96]. It can realize the on-site test of SARS-CoV-2 on cold-chain food using simple equipment. The test time varies from 40 to 70 min when excluding the time for RNA extraction ^[52][53][97,98]. Combined with RT-LAMP technology, the CRISPR-Cas system can achieve an LoD of 10 copies/μL. Most CRISPR-based SARS-CoV-2 detection methods use the Cas12 enzyme to specifically recognize the virus sequence ^[54][99]. In addition, the all-in-one dual CRISPR-Cas12a analysis system does not need a pre-amplification step and it improves the sensitivity of the assay by using double CRISPR RNA. It can detect 1.2 DNA targets and 4.6 RNA targets in 40 min. The system can be developed as a one-step test platform without the need for cDNA preparation which has the potential for SARS-CoV-2 detection on cold-chain food ^[55][100].

Microfluidic chips integrate various small-scale laboratory functions on a single chip to complete the steps in traditional laboratories ^[56][101]. It uses a small number of reagents and samples to obtain accurate test results in a short time, and is especially suitable for the rapid detection of SARS-CoV-2 on cold-chain food. Recently, paper-based microfluidics, centrifugal chips, wearable microfluidic devices, and digital nucleic acid detection chips have been proposed for pathogen testing and disease screening ^[57][102]. For instance, the IDNOW^® instrument proposed by Abbott™ in the United States can detect a positive sample. The product has received an emergency use authorization (EUA) from the U.S. Food and Drug Administration (FDA). The instrument weighs only 3 kg and is portable and suitable for POCT ^[58][59][103,104].

Whole-genome sequencing (WGS) is an effective tool to comprehensive understand SARS-CoV-2. The assay belongs to high-throughput sequencing, or next-generation sequencing, which is a culture-free, unbiased, direct extraction of DNA or RNA from clinical samples ^[3][4][52,53]. However, the operation steps of WGS are relatively complex and the operation technology requirements are high. The RNA can be extracted using the kit and whole-genome sequencing performed on an instrument (e.g., Illumina iSeq 100). Peculiarly, metagenomics is a high-sensitivity pan-pathogen assay for the discovery of novel pathogens and infectious disease diagnosis, which is applied in the simultaneous and rapid detection of SARS-CoV-2 ^[60][105]. Because the whole gene sequencing requires a professional operator, complex sample pretreatment, and long-term periods, other methods are usually combined to generate test reports.

SARS-CoV-2 has a wide mammalian host range, including minks, cows, white-tailed deer, dogs, domestic cats, swine, lions, etc. ^{[61][62][63][64]}[106,107,108,109]. Some of these animals may serve as viral carriers once they are made into food-related products ^[65][110]. Immunological tests can directly detect the antigen biomarker of SARS-CoV-2 and can be used to determine whether food (e.g., animal products) is infected or contaminated by the virus. The immunological methods that mainly include antigen tests, serology tests, and cytokine storm diagnoses can be used for SARS-CoV-2 detection for these animal foods. For plant foods, that cannot undergo an immune response to produce antibodies, serum antibody immunological testing and cytokine storm diagnosis cannot be applied to the detection of cold-chain food unless the plant food is contaminated with the body fluids of an infected person ^[66][111]. In the following sections, rwesearchers will describe each of the above methods in detail.

Antigen tests are the main immunological method for food quarantine. The structural proteins, such as the spike glycoprotein (S), envelope protein (E), membrane protein (M), and nucleocapsid protein (N) of SARS-CoV-2 are the primary antigens used for the immunological test. Among the antigen immunological test methods (Figure 1B), ELISA is a highly sensitive immunological experimental technique that combines an antigen, antibody-specific reaction, and high-efficiency enzyme catalysis on the substrate. This detection method has high sensitivity and low difficulty in carrier standardization, but the detection steps are more cumbersome and easier to contaminate ^[67][112]. ELISA kits can test multiple samples in a single run; however, they lack point-of-care applicability and the non-specific binding of antibodies or antigens to the plate may lead to false positive results ^[68][113]. In addition to ELISA, the detection of trace S-protein (S1 subunit) for real-time SARS-CoV-2 detection is currently a known method that can be well applied to cold-chain foods. The S-protein particles can be attracted to the surface of the sensor and captured by the antibody within 20 s, which meets the real-time detection requirements on-site. The linear range is wide and covers the possible range of the concentration of S-protein on the food surface. The developed ultra-low LOD strategy has shown great advantages in the detection of virus markers with low concentrations of cold-chain foods ^[23][25]. The single sensor device can act as a disposable chip and its cost is estimated to be 1 US dollar. The operations of the device are relatively simple and can be operated by non-technical personnel ^[69][114]. Thus, the S-protein detection platform meets the requirements of rapid response, lower detection limit, high specificity, friendly operation, and low cost, which provides a promising solution for SARS-CoV-2 detection on cold-chain food ^[70][115].

Besides, the biosensor equipped with a chemical or biological receptor (e.g., antibody) can specifically interact with the target analyte showing a quantitative signal of the recognition process ^[71][116]. Compared with traditional laboratory methods, biosensors can provide a cheap, sensitive, rapid, miniaturized, and portable platform for SARS-CoV-2 detection which is promising for the on-site detection of cold-chain food contamination. Several studies have proved biosensor technology is convenient in SARS-CoV-2 S protein detection based on a bioelectrical identification assay ^[72][117]. The biosensor can detect S protein within 3 min with a LOD of about 1 fg/mL, and no cross-reaction with SARS-CoV-2 nucleocapsid protein was found. The portable readout system of the ready-made biosensor platform can be controlled by a smartphone or a tablet computer ^[73][118]. The high sensitivity, rapidness, and simplicity of biosensors make it a great advantage in the detection of viruses on cold-chain food. It can be easily controlled and has strong practical applicability to test contaminations on cold-chain food in factories or customs. However, some biosensors involve the use of enzyme reagents, and the biosensor technology is still in the development stage ^[74][119].

Once the body is infected, the living organisms will produce specific antibodies, such as anti-SARS-CoV-2 IgM and IgG. Though animal food is frozen or transported in the cold-chain, the antibodies from an earlier infection can be preserved. The safety risk of infected COVID-19 is relatively low by eating cold-chain animal food and has not been studied to date. However, these serum antibodies can be used as the immunological target, especially in cold-chain animal food with a low viral load. Compared with nucleic acid testing, blood samples for antibody serology testing are easier to obtain, which greatly reduces the risk of infection of medical staff during specimen collection and testing, and makes it easier for primary laboratories to carry out screening work. For instance (Figure 1C), luciferase immunosorbent assay (LISA) is an easily and rapidly developed semi-quantitative method and is appropriate for detecting specific antibodies from cold-chain animal food ^[31][78]. Compared with enzyme-linked immunosorbent assays, DNA-assisted nanopore sensing assay can reliably quantify SARS-CoV-2 antibodies with high accuracy, wide dynamic range, and the potential for automated detection on cold-chain food ^[29][76]. Though modified with probe DNA to label IgG or IgM antibodies, the nanopore sensor can quantify the probe DNAs when thermal dehybridization of gold nanoparticles (AuNPs) probe DNAs was performed. In addition, surface plasmon resonance (SPR) biosensors assay adopts the optical detection method. Indirect aggregation can be used for virus detection by modifying targeted molecules on the virus surface. If you do not consider the portability of the SPR device, it may be the most promising technique for cold-chain food quarantine ^[28][75].