3.3 Sequencing
This stage involves loading your DNA library into the sequencing machine and running the appropriate sequencing software to determine the sequence of DNA. The basic unit of DNA is a
Two approaches to WGS are commonly used:
- Second-generation (short read) sequencing produces short reads (DNA fragment sequences) (50–500 base pairs) by running many reactions at once.
- Third-generation (long read) sequencing uses newer methods to read much longer stretches of DNA.
Each technology has its own advantages and limitations; Table 3 has a detailed comparison of them.
| Point of comparison | Short-read sequencing | Long-read sequencing |
|---|---|---|
| Sample commercial machines | 454 Roche, Ion Torrent, Illumina | PacBio, Oxford Nanopore |
| Read length (i.e. the size of the DNA fragments – longer reads simplify post-processing and analysis) | Short pieces of DNA (about 50–500 base pairs long) | Long pieces of DNA (up to thousands or even millions of base pairs long) |
| Accuracy | Currently has higher accuracy, generating higher-quality sequences | Lower accuracy but better resolution of complex regions |
| Applications | Finding genetic mutations Microbiome and AMR studies Targeted gene testing Outbreak and AMR surveillance; however, may miss structural variations or struggle with repetitive regions | WGS Complete genome assembly Detecting large genetic changes, gene arrangements and mobile elements Plasmid detection Real-time outbreak and AMR surveillance Mobile/field-based testing |
| Cost and speed | High initial equipment cost Lower per-sample cost when run at scale Best for large batches May have slower turnaround in low-volume setting | Lower upfront equipment cost Higher per-sample cost Ideal for small batches or rapid response Faster turnaround for fewer isolates |
| Institutional considerations | Larger machines suitable for centralised facilities High throughput and cost-efficient when used fully | Portable machines (e.g., MinION) Enables decentralised or field-based sequencing Real-time sequencing, quick turnaround time |
| Number of isolates that can be sequenced in one run | Depends on the platform used but typically ranges from around 80–100 isolates on a MiSeq up to 700–900 on a NextSeq, and as many as 2500–3500 on a NovaSeq | 12–24, depending on the flow cell and kit being used |
For short-read sequencing, Illumina machines are the most widely used machines in all genomics work. There is a wide range of Illumina machines (Figure 6), with varying prices and specs. As an example, a SeqAfrica partner laboratory in Ghana purchased an Illumina Nextseq 1000, including freight, installation and training for around £186,000 (2023).
Video 1 shows how an Illumina workflow functions.
Oxford Nanopore Technologies (ONT) produces the most used long-read sequencing devices. A Nanopore MinION Mk1D pack including sequencing consumables and five MinION Flow Cells cost around £3600 in 2025. As seen in Figure 7, the MinION is a portable real-time sequencing device.
The MinION device needs to be connected to a PC or laptop (for field-based analysis) to run and collect data. Oxford Nanopore Technologies’ website includes a video of how nanopore sequencing works [Tip: hold Ctrl and click a link to open it in a new tab. (Hide tip)] .
For both types of sequencing, the companies produce a variety of machines, each with different levels of accuracy and potential throughput.
3.2 DNA isolation and library preparation
