Faster analysis of cancer-associated variants using a unique targeted nanopore sequencing approach

Targeting specific gene regions with known or suspected association with cancer enables faster analysis, higher depth of coverage, and reduced sequencing costs compared to whole-genome sequencing. Whereas most sequencing technologies require time-consuming upstream, lab-based enrichment procedures, nanopore sequencing offers a unique, on-device enrichment method — adaptive sampling — which does not require any additional sample prep. Since nanopore sequencing is performed in real time, as DNA passes through a nanopore, it is possible to select for DNA fragments of interest and eject off-target regions.

One major advantage of target adaptive sampling … is that the user only needs to specify the coordinates of the target area, and no prior sample prep is required1

Nakamura et al. in Japan developed an adaptive sampling workflow to ‘precisely detect’ both genomic and epigenomic variants in suspected hereditary cancer blood research samples1. Cancer cells are characterised by the accumulation of genetic and epigenetic variants, and at  least 30% of cancers have a known pathogenic structural variant (SV) — genomic aberrations >50 bp — which is used in diagnosis or treatment stratification2. Using  short-read sequencing (SRS), read lengths are limited by fragmentation and amplification, meaning that many SVs cannot be captured within a single read, and remain undetected. With direct nanopore sequencing of native DNA, there is no limit to read length — long nanopore reads can span entire SVs in single reads and retain epigenetic information.

This striking method can be flexibly executed by simply specifying the target regions computationally1

Applying adaptive sampling for software-level enrichment during sequencing on a GridION, Nakamura et al. analysed putative pathogenic SVs — those which disrupt the coding sequence. Using traditional technologies, including SRS, many repetitive and complex SVs can only be inferred or partially identified. Using targeted sequencing of long nanopore reads, the team successfully characterised suspected SVs — with precise breakpoint junctions. One example included sequencing and analysis of a familial adenomatous polyposis research sample, in which they were able to characterise a complex SV in the key tumour suppressor gene APC, comprised of reciprocal inversions and a 130 kb deletion. Novel SVs were also reported, including SINE-R/VNTR/Alu (SVA)-derived mobile insertion elements which are ‘notoriously difficult to detect with existing short-read sequencing platforms1.

An important advantage’ of nanopore sequencing, the authors highlighted, is that epigenetic modifications such as methylation can be obtained with no additional library preparation. Long nanopore reads also enabled methylation to be phased according to haplotype. Implementing a workflow to automatically detect allele-specific methylation targets, the team were able to reveal an epigenetic mutation in the MLH1 gene from a Lynch syndrome research sample. Previous multi-gene panel testing  using traditional SRS technology had failed to detect germline variants within the sample. Nanopore sequence analysis revealed that the promoter region of one allele of MLH1 exhibited hypermethylation, suggesting that gene expression was reduced in one allele (Figure 1). Their analysis suggested that this represented a novel epigenetic mutation, characterised 'for the first time' within this research sample.

Figure 1. Direct methylation detection with nanopore sequencing of native DNA. Long nanopore reads enabled haplotype phasing: methylation was specifically increased in haplotype 2. Image taken from Nakamura et al. and available under Creative Commons license (

The team also characterised single nucleotide variants (SNVs), highlighting novel SNV findings, including the identification of a potential splicing variant in the EPCAM gene, ‘registered as pathogenic’, in a suspected Multiple Endocrine Neoplasia type 2 (MEN2) research sample and polymorphisms in the RET proto-oncogene which has been previously associated with MEN2. Describing how discarded reads from the adaptive sampling workflow can be used to accurately genotype common single nucleotide polymorphisms (SNPs), they calculated a polygenic risk score for three cancer types, highlighting that with nanopore technology, ‘a single platform can simultaneously capture the genomic and epigenetic status of monogenic disease-causing genes as well as polygenic effects’.

‘An important advantage of [nanopore sequencing] is that epigenetics modifications such as methylation can be obtained for each sequence read and position’1

In a separate study, Filser et al. in France used adaptive sampling on the MinION to test the potential of the method to capture cancer-associated variants3. Using a breast cancer research sample, the group aimed to test the potential of the technology to ‘better characterise and classify’ an SV in a shorter timeframe and at lower cost than currently used methods. Noting that the ‘functional consequences [of SVs] can be difficult to assess … as the precise resolution of their sequence is often impossible with short-read next generation sequencing’, they assessed the potential of nanopore sequencing to accurately resolve a germline duplication event, previously classified as a ‘variant of unknown significance’ in BRCA1, a tumour suppressor gene.

‘By computationally selecting molecules during the sequencing process, the process [is] simpler, faster and easily adaptable.’3

Motivated by the knowledge that ‘long reads do not suffer from the same mapping issues as short reads in repetitive regions and permit the detection of SV at a base-pair resolution’, they used adaptive sampling to characterise a tandem duplication event in exons 18–20 of the BRCA1 gene, and precisely identified SV breakpoints located in two Alu repetitive elements sharing 74% identity. The long nanopore reads revealed that the duplication introduced a premature stop codon, thus altering the reading frame. Their results supported the hypothesis that this SV was mediated by non-allelic homologous recombination.

The authors explained that since SRS cannot differentiate between a potentially pathogenic tandem duplication event and a duplication of unknown significance, they would normally use a cDNA analysis technique  , which has an ‘average turn-around time of 2 months’. Using the ‘extremely flexible’ nanopore adaptive sampling approach to target SVs, Filser et al. demonstrated the capacity to identify a tandem duplication in ‘less than 10 days’, and concluded that ‘nanopore sequencing coupled with adaptive sampling was demonstrated to be an effective, reliable and fast long-read sequencing technique’.

Read the paper (Nakamura et al.)

1. Nakamura, W. et al. A comprehensive workflow for target adaptive sampling long-read sequencing applied to hereditary cancer patient genomes. medRxiv June (2023). DOI:

2. van Belzen, I.A.E.M. et al. Structural variant detection in cancer genomes: computational challenges and perspectives for precision oncology. NPJ Precis. Oncol. 5(1):15 (2021). DOI:

3. Filser, M. et al. Adaptive nanopore sequencing to determine pathogenicity of BRCA1 exonic duplication. J. Med. Genet. Epub ahead of print (2023). DOI: