A Two-color Single-molecule Sequencing Platform and Its Clinical Applications

Abstract

DNA sequencers have become increasingly important research and diagnostic tools over the past 20 years. In this study, we developed a single-molecule desktop sequencer, GenoCare 1600 (GenoCare), which utilizes amplification-free library preparation and two-color sequencing-by-synthesis chemistry, making it more user-friendly compared with previous single-molecule sequencing platforms for clinical use. Using the GenoCare platform, we sequenced an Escherichia coli standard sample and achieved a consensus accuracy exceeding 99.99%. We also evaluated the sequencing performance of this platform in microbial mixtures and coronavirus disease 2019 (COVID-19) samples from throat swabs. Our findings indicate that the GenoCare platform allows for microbial quantitation, sensitive identification of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus, and accurate detection of virus mutations, as confirmed by Sanger sequencing, demonstrating its remarkable potential in clinical application.

1. Introduction

Beginning in 1990, the National Human Genome Research Institute of the National Institutes of Health spent 14 years assembling the first version of the human genome [1]. After this landmark achievement, different sequencing techniques have been developed and commercialized to reduce the cost and increase the efficiency [2–9]. The resulting high-throughput and low-cost genomic sequencing techniques, especially next-generation sequencing (NGS) based on clonal amplification and sequencing-by-synthesis (SBS), have revolutionized biomedical research and clinical diagnosis [10–13].

Single-molecule sequencing (SMS) technologies have also been developed. The key feature of this method is the ability to conduct sequencing without amplifying target DNA, thereby avoiding the errors and biases introduced by amplification in NGS platforms [14]. Based on different sequencing principles, the three types of SMS techniques are as follows: true SMS (tSMS; Helicos BioSciences), employing virtual reversible terminator chemistry [5,15]; single-molecule real-time (SMRT) sequencing (Pacific Biosciences), which uses a zero-mode waveguide to monitor fluorescence signals emitted by single DNA polymerase extension [9,16]; and nanopore sequencing (Oxford Nanopore Technologies), which measures the current change when DNA is translocated through protein pores [8,17]. Although SMRT and nanopore sequencing can generate long reads (> 10 kb), their applications are limited by low read numbers, high reagent costs, and complicated sample preparation processes. Particularly for clinical applications such as non-invasive paternal test (NIPT) and pathogen detection by metagenomics sequencing, their costs and lengthy protocols make them less attractive compared to NGS sequencing.

Here, we developed an SBS-based SMS platform, which demonstrates high read throughput and accuracy without requiring amplification. In contrast to tSMS, our GenoCare platform is based on two-color chemistry, which improves reaction efficiency, reduces sample-to-data time, and lowers the reagent costs. The 16-channel flow cell could deliver more than 300 million reads per cell, which is one to two orders of magnitude higher than SMRT and nanopore sequencing. The sample preparation and sequencing can be completed in 24 h. Recently, the application of GenoCare in NIPT and tuberculosis studies has been reported [18,19]. For NIPT, SMS improves the detection of common fetal aneuploidies by reducing the GC bias introduced during library preparation and sequencing [19]. For Mycobacterium tuberculosis mutant re-sequencing (250× genome coverage), it has been found that mutations in the gyrA and gyrB genes are the main mechanisms of gatifloxacin resistance [18]. To investigate the performance of the GenoCare platform in pathogen detection, we sequenced the Escherichia coli genome, microbial mixtures, and coronavirus disease 2019 (COVID-19) patient samples from throat swabs. The E. coli sequencing was finished within 15 h with an average read length of 53 bp and consensus accuracy exceeding 99.99%. Microbial mixture sequencing showed the quantitative response of the sample concentrations.