A full genome tiling array enhanced the inspection and quarantine of SARS-CoV-2

The primary tool that is the most widely used viral nucleic acid detection method is the real-time PCR (RT-PCR) method [1, 2]. With its high sensitivity and specificity, RT-PCR is considered a gold standard for diagnosing Covid-19 (https://​apps.​who.​int/​iris/​handle/​10665/​331329). But as the Covid-19 pandemic continues, mutations and virus evolution bring new challenges to the method. Not only for the possibility of mutation-associated false negative [3] but also for the lack of ability to identify mutations and variants essential for making policy decisions regarding outbreak control. Next-generation sequencing (NGS) technology is an alternative virus nucleic acid detection method. This method can obtain the virus genome sequence, and by comparing it with a reference genome sequence, mutations and haplotypes can be identified [4]. We can track the origin and evolution by identifying mutation patterns or variants of the virus. Even if the costs have come down and the difficulty of protocols has eased as a legacy of the Covid -19 pandemic, however, the NGS method is limited by its relatively high cost, time-consuming and complex experimental procedures, and bioinformatic requirements for data analysis. Therefore, using NGS sequencing in large-scale clinical detection is challenging, especially on the front lines of pandemic control, such as in customs inspection and quarantine.

Microarray technology is a rapid and high throughput molecular biology detection tool. The detection rate of the re-sequencing chip can reach 96–99%, and the consistency with Sanger sequencing can reach as high as 99.99% [5]. Wang et al. have developed a re-sequencing chip detection method for more than 100 microorganisms [6]. Guo et al. have designed a microarray for detecting SARS based on Affymetrix high-density gene chip technology [7, 8]. The RPM v.1 and PathChip were respiratory pathogen arrays designed to detect various respiratory pathogens [9, 10].

Our study uses a previously reported tiling array chip to sequence the whole genome of SARS-CoV-2 [11] so that the informative mutations can identify the variants. With the ability of rapid and inexpensive full viral genome re-sequencing, this method is expected to be a tool for mutation monitoring and virus source tracing during daily inspection and quarantine in the epidemic.

This study used a 3 mm × 3 mm silicon-based high-density in situ synthetic tiling array chip, which contains over 250 k probes for sequencing the full genome sequence of SARS-COV-2. The testing workflow consists of two main sections. The first section is library construction, including RT-PCR (4 h) and DNA fragmentation (45 min). The second section is chip assay, including hybridization (2 h), staining (30 min), scanning (8 min per chip), and data processing (3 min per chip). The workflow goes from sample to sequence and can be finished in one day with a throughput of as high as 96 chips in parallel.

The next-generation sequencing (NGS) method was used in the previous study to verify base calling accuracy. The results indicated that the average accuracy of chip-based re-sequencing could be greater than 99.9% over 95% of ~ 30,000 bases SARS-CoV-2 genome [11]. To further validate the assay and certificate the detected mutations, 6 clinical samples with different Ct values were tested. All these samples successfully detected the whole SARS-CoV-2 genomes with relatively high coverages ranging from 99.68 to 99.96% (Table 2). Three mutations that existed in all 6 samples were confirmed by Sanger sequencing. The result showed that all these mutations called by our tiling array were consistent with Sanger sequencing (Fig. 2).

Variants can potentially affect the transmission, disease severity, diagnostics, therapeutics, and natural and vaccine-induced immunity [16]. The emergence of highly transmissible SARS-CoV-2 variants of concern (VOCs) such as Delta and Omicron has given rise to massive outbreaks in many countries. It is critical to discriminate against different variants caught at entry and exit ports to support policy decisions. Compared with PCR technology, a full genome tiling array can avoid false negatives caused by mutation occurring at primer binding regions [17]. It can also provide genome sequences and mutations of the virus, which is of great significance for virus traceability, virulence, and vaccine evaluation. Sanger sequencing [18] and Next-generation sequencing (NGS) are mature technology at present. The WHO has recommended that countries sequence at least 1% of their SARS-CoV-2 positive samples to detect emerging VOCs and significant mutations in the virus. Yet, the cost and complexity of workflow make Sanger sequencing and NGS challenging to be widely applied in the front line of anti-epidemic, such as entry and exit ports. For the tiling array, the processing duration from RNA virus to sequencing data is controlled within one day. Whole genome FASTA and FASTQ files are output using customized software based on a bioinformatics protocol of two base calling methods [12]. According to the Sanger sequencing results in this study and the previous reports, this assay accompany the base calling algorithms has ability to discover mutations with a genome-wide accuracy of at least 99.5% [12]. More importantly, this method is highly cost-efficient with $30 per sample and high throughput with 96 parallel testing abilities. Unlike the high-throughput amplicon sequencing[19], barcoding which takes at least 2 h with laborious benchwork is not required for this method. The method in this study is less dependent on equipment and professionals than NGS-based methods. The required equipment includes hybridization ovens, PCR machines, and a chip reader. The operators only need to master the basic operation skills of molecular biology experiments. Of course, like RT-PCR, sequencing results are mostly affected by the viral load in the samples. According to the detection results of clinical samples with known CT values, the coverage of chip sequencing can reach higher than 99% for samples with CT values below 31 so that accurate lineage assignment can be obtained using Pangolin. Furthermore, an updated version of ARTIC multiplex PCR primer sets can improve accuracy at a new emerged variant like omicron [3].