ApplicationsInvestigations

Single nucleotide variants (SNVs) and phasing

Single nucleotide variants (SNVs) have been widely studied for their associations with phenotypic variation and disease; they are also used to phase haplotypes. However, the use of traditional sequencing technology requires PCR, limiting SNV detection to regions amenable to amplification, and short reads make phasing challenging to resolve. With nanopore sequencing, PCR is not necessary, revealing SNPs in regions inaccessible to other technologies, and phasing is greatly simplified.

Phase SNVs and resolve compound heterozygosity using long sequencing reads

Perform real-time analysis with high precision and recall

Scale to your requirements with a range of nanopore sequencing platforms

Subscribe Get started Talk to us

Introduction

Enhance variant calling and phasing with long sequencing reads

Array-based and short-read sequencing analyses of SNVs have shed light on the association between SNVs and the inheritance of complex diseases. However, data from such technologies is only available when a run is completed, which can prevent a rapid response in situations where speed may be vital, such as in outbreak investigations, forensic analysis, and clinical research. Real-time sequencing and analysis with Oxford Nanopore Technologies provide rapid sample-to-genotype, enabling an immediate response. The technology is also scalable: high-throughput variant calling in larger genomes or in multiplexed samples can be performed with GridION and PromethION; the portable MinION is ideal for variant calling in smaller genomes, and for targeted detection, both in the laboratory and in the field.

Assigning a variant to the maternal or paternal chromosome is important for understanding inheritance patterns, mosaicism, and parental origin of de novo mutations, for example. To directly resolve the haplotype of two heterozygous SNPs, they both need to be present within the same sequencing read. This is inherently challenging with short sequencing reads. Nanopore long reads provide sequence context and enhance the phasing of variants (Figure 1).

Figure 1: Compared to phasing with short-read sequencing or array-based data, phasing of single nucleotide variants (SNVs) is enhanced with long nanopore sequencing reads, as variants are more likely to be located within the same read.

Figure 2: The latest R10.4.1 data for accuracy measured as F1 (harmonic mean of precision and recall) for small variant calling, using nanopore sequencing data for the human genome (HG002 cell lines) at 20x, 30x and 60x (Kit V14 400 bps, basecalling models of High Accuracy, HAC and Super Accuracy, SUP, using Guppy v6.3.2). Variant calling was then performed with the latest version of the Clair3 variant caller, and variants were compared against the Genome In A Bottle consortium’s HG002 truthset (v4.2.1). SNVs and Indels are represented with solid colours, while Indels in CDS regions are displayed with shaded colours

Nanopore sequencing enables genome-wide variant calling with high precision and recall

Traditional short-read-based methods for SNV calling require PCR, which can introduce bias, and prevents investigation into genomic regions that are not amenable to amplification. With Oxford Nanopore, native DNA strands can be sequenced, negating the requirement for PCR; this enables greater breadth of genome coverage for SNV calling and unambiguous mapping in repetitive regions, as well as access to methylation information from the same sequencing run as standard.

Genomic variants can be characterised with high accuracy using nanopore sequencing data. This is summarised by F1 score, the harmonic mean of precision and recall, in Figure 2, where we display the latest Oxford Nanopore benchmarking metrics for SNP and indel calling in the human genome. The development and optimisation of variant calling tools is a very active area of research. Different bioinformatics tools for SNP calling show variation in performance, and therefore it is important to optimise filtering parameters of the analysis workflow to achieve the best sensitivity and specificity.

Case study

Phasing analysis of lung cancer genomes using long nanopore reads

'[With Q20 chemistry] it is now realistic to use long read sequencers to systematically analyze a wider range of cancerous mutations'

Sakamoto et al. 2022

Approximately 20–30% of lung cancer patients remain undiagnosed with respect to their cancerous mutations; variant detection is particularly pressing in cases where an effective treatment remains elusive. To investigate the potential of long nanopore reads for cancer variant analysis, Sakamoto et al. sequenced cancer cell lines and lung cancer research samples and performed integrated analysis of phased genomic, epigenomic, and transcriptomic aberrations. They were able to phase data into long phase-blocks ,revealing allele-specific information; this phased data was also then paired with transcriptomic data. Direct methylation calling, in conjunction with the long phase blocks, allowed allele-specific methylation to be derived.

Authors identified a unique class of complex structural variant, consisting of copy number changes, inversions, and deletions, which could not be characterised using short-read sequencing. Phasing cancer-associated mutations revealed valuable information regarding allele-specific methylation and also regions in which mutations were unevenly distributed to either haplotype. The authors concluded that these mutations may point to the causal mechanism of disease in clinical research samples where this is unknown.

Read the paper

View our workflow for calling and phasing SNVs in the human genome

Sequencing workflow

How can I call and phase SNVs/SNPs with nanopore sequencing?

To perform genome-wide small variant calling with nanopore technology, we recommend sequencing a whole human genome on the PromethION, or sequencing on the MinION or GridION for targeted analyses. Libraries can be prepared using the Ligation Sequencing Kit, for high throughput and long reads, without amplification.

We recommend wf-human-variation for calling single nucleotide variants and indels. However, other analysis tools are also available from the Community.