Adaptive Sampling: reimagining targeted sequencing

Targeted sequencing — the selective sequencing of specific regions of interest rather than entire genomes — is a critical technique across numerous applications, including clinical research, pathogen surveillance, and species identification. Traditionally, the methods used to achieve targeted sequencing require enrichment or depletion via library prep before the sample reaches the sequencer. These library prep techniques, whilst effective, are often prone to PCR bias, complex, costly, and time consuming.

What if targeted sequencing could truly mean targeted sequencing — the enrichment of targets of interest, or depletion of off-target regions, during sequencing itself, with no need for special library prep steps? In this Nanopore Know-How blog, we introduce Adaptive Sampling: a method unique to Oxford Nanopore sequencers that enables just that.

Why use targeted sequencing?

If you only require data from a fraction of a genome, or one genome from a mixed microbial community, sequencing the entire sample may provide insufficient data for characterising these targets. Additionally, extra sequencing runs would increase costs, extend turnaround times, and lead to more complex data management.

This is where targeted sequencing shines. Instead, by focusing sequencing on targets of interest, you achieve high depth of coverage of targeted regions via a faster and more efficient workflow than whole-genome sequencing.

How is targeted sequencing typically performed?

Traditional targeted methods utilise library prep techniques to enrich specific regions before sequencing. A commonly used approach is targeted PCR, also known as amplicon sequencing, which can offer a simple and cost-effective workflow. However, the trade-offs are that PCR leaves behind regions that cannot be amplified, such as GC-rich and repetitive sequences, restricts analysis to fragment lengths that can be processed by the polymerase, introduces bias, and erases epigenetic modifications. When combined with legacy short-read sequencing technology, read lengths are further reduced, hindering the analysis of large regions of interest such as structural variants (SVs).

Hybridisation capture is another method of targeted sequencing, typically used for the enrichment of wider target panels. This technique also relies on PCR, and requires a complex, lengthy workflow. Depletion techniques can also be performed to remove unwanted DNA, such as host material in microbial samples, but these again require additional sample prep steps.

Why is Adaptive Sampling different?

Adaptive Sampling changes the game. With this innovative method developed by the Nanopore Community^1–4, you can perform target enrichment or depletion entirely during sequencing, on the device itself. The software-based method makes it possible to perform targeted sequencing without sacrificing long reads or native DNA. This means you can target sequencing to previously inaccessible regions, including complex SVs, long repeats, and GC-rich sequences — and detect native DNA modifications alongside genomic variants, in one go.

Adaptive Sampling is a fast, simple technique that requires no PCR, no time spent designing and ordering probes or waiting for them to be synthesised, and no extra wet-lab steps.

How does Adaptive Sampling work?

Your Oxford Nanopore sequencing device is controlled by the easy-to-use software MinKNOW, and this is what you use to set up your sequencing run, including Adaptive Sampling. After specifying whether you wish to enrich for or deplete your specified target sequences, all you need to provide is a BED file containing the coordinates for these targets — more on this later — and a reference for alignment. Then, your targeted sequencing run is ready to start.

The targeting itself is made possible by the real-time nature of Oxford Nanopore sequencing and data analysis, which allows a DNA strand to be basecalled whilst it is still passing through a nanopore. During Adaptive Sampling, the sequence at the beginning of each DNA strand is compared against your list of targets. If the sequence matches a target of interest or is not a region to be depleted, the strand is ‘allowed’ by the software to continue sequencing. If it is not a target of interest or is listed for depletion, the DNA molecule is ejected from the nanopore so that it can begin sequencing another molecule.

This process significantly increases the depth of coverage for your targets of interest across a sequencing run. Adaptive Sampling can be used with any Oxford Nanopore sequencer — either on MinION Flow Cells on a MinION or GridION, or PromethION Flow Cells on a PromethION.

You can use Adaptive Sampling to enrich for large, complex sets of targets. You can target huge regions with ease — even entire chromosomes². Meanwhile, with ‘depletion’ mode, you can perform host depletion, such as of human DNA from microbial samples, or deplete abundant microorganisms from a metagenomic sample to capture rarer or understudied microbes of interest^2,3.

With this technology, you can enrich panels of targets across a genome, such as disease-associated genes, simply by specifying their coordinates in a BED file. We call these 'digital panels'. You can use one of our ready-to-use BED-file-defined digital panels or construct your own panel of targets using our intuitive browser-based tool.

What ready-to-use digital panels do Oxford Nanopore offer?

We offer a range of ready-to-use research-use-only (RUO) digital panels, spanning multiple research applications. Our current portfolio includes:

• Hereditary Cancer Panel*: targets 258 genes associated with hereditary cancer risk
• Pharmacogenomics Panel*: targets 375 genes associated with pharmacogenetic effects, including full resolution of the CYP2D6 gene
• Pan-Cancer Panel*: targets 723 genes associated with cancers. This RUO digital panel is used in our targeted tumour methylation profiling workflows
• Pan-Haematological Cancer Panel*: targets 144 genes associated with haematological cancers

How do I create a custom digital panel?

Using our online Digital Panel Designer, you can create, modify, and update your own digital panels to target any regions of interest for your research. Generating your own panel is as simple as providing the gene names or genomic coordinates of the regions you want to enrich. We recommend targeting 0.1–5% of your reference genome.

With no need for the lengthy process of probe synthesis or complex probe panel optimisation, you can easily update digital panels by editing your BED file of coordinates in a matter of minutes as your research evolves. You can also use the Digital Panel Designer to edit our ready-to-use digital panels to suit your research goals — for example, by adding or deleting target genes.

What about library prep?

No special library prep is required for Adaptive Sampling. Your whole DNA sample is prepared for Oxford Nanopore sequencing in the same way as for a whole-genome experiment. The PCR-free Ligation Sequencing Kit only takes around one hour to prep native DNA for sequencing, preserving long fragments and base modifications. You can also prepare samples for multiplexed sequencing on a high-output PromethION Flow Cell using the Native Barcoding Kit.

The Adaptive Sampling workflow utilises more input library than a standard Oxford Nanopore workflow, ensuring that plenty of on-target DNA molecules are available for sequencing. Light fragmentation of high molecular-weight DNA helps to further improve data outputs and can also help to increase enrichment.

How are the Nanopore Community using Adaptive Sampling?

Scientists are harnessing the unique benefits of Adaptive Sampling to enhance their research across a wide range of applications. Researchers have shown future potential of the technique to enable intraoperative classification of central nervous system tumours⁵ and cancer-associated pathogenic variants across the complete MSK-IMPACT⁶ panel without the need for lengthy hybridisation capture⁷. Adaptive Sampling has also been used to genotype all known neuropathogenic short tandem repeat expansions, plus pharmacogenomic targets, in a single sequencing run⁸, showing its potential to tackle the ‘diagnostic odyssey’ faced by many families. Going beyond the human genome, the method has been used to target DNA from the foodborne pathogen Staphylococcus aureus without the need for culturing⁹, and Plasmodium falciparum parasite DNA from blood research samples¹⁰.

With this unique technology, the Nanopore Community continues to push the boundaries of what is possible with targeted sequencing.

How will you use Adaptive Sampling?

Learn more in our Adaptive Sampling getting started guide

View Adaptive Sampling documentation

*Research-use only. Not for use in diagnostic procedures.

Oxford Nanopore Technologies products are RUO. Products labelled/branded as OND may be RUO or may be regulated as in-vitro diagnostic devices in some jurisdictions.