The Nanopore Community Meeting 2022: more than 4000 people registered to hear the latest in nanopore sequencing

Day two of this year’s Nanopore Community Meeting saw the return of a live audience in New York, as well a huge online attendance via the virtual platform. The first live day of the conference saw speakers presenting across a mix of stages including Plenary, Breakout, Showcase Stage, Mini Theatre, Poster and Spotlight Sessions. The day started with the hugely popular ‘data for breakfast’ on the Innovation Stage and below is a summary of the highlights of the day.

Welcome

Spike Willcocks, Chief Strategy Officer and Co-Founder, started the day with a reminder of the long journey from the early years of Oxford Nanopore Technologies. He went on to showcase how much the Community has delivered since the MinION launched in 2014, how the Nanopore Community now has more than 39,000 registered members and can be found in more than 120 countries worldwide. Importantly this Community has also published more than 3500 peer reviewed publications.

Spike went on to discuss how the Nanopore Community has, and is, breaking new frontiers. Since the first published nanopore data in 2014, nanopore sequencing has been brought to the classroom, the International Space Station, to farmers protecting their crops in Africa and many more places. Spike finished by thanking the community for their inspiration and asking what will the Nanopore Community discover next?

Expanding studies of global genomic diversity with complete, telomere-to-telomere assembly of diploid genomes – Karen Miga

Karen opened her presentation by celebrating the fact that the human genome sequence was finally completed and published in Science earlier this year, introducing ~200 Mb of previously unresolved sequence and providing a more accurate reference. This will enable new science exploring “regions we thought were intractable”. Karen also noted that “short-read data is incomplete” when it comes to studying the epigenome – and that only with base modification calls “we can study the genome end to end”; “this is what it means to be comprehensive”.

However, one complete genome of an individual with European ancestry is not enough and Karen now directs the Human Pangenome Reference Consortium, a multi-centre programme aiming to improve representation of global genome diversity, with over 350 diverse diploid references. To achieve this feat, 30x of ultra-long nanopore whole-genome sequence data will be obtained for each individual human diploid genome, and the tool Verkko will be central to assembling complete, diploid genomes.

Karen then shared her ‘wish-list’: longer high accuracy (Q30) reads; one machine under one roof; and reducing per-genome cost. Regarding the first two items on her list, Karen shared her team’s latest cutting-edge efforts to obtain and incorporate duplex data, producing longer, high accuracy nanopore reads (mean Q-score 30.1). From one PromethION device they can now incorporate duplex data, ultra-long read data, and Pore-C (chromatin conformation capture data, using the tool GFAse), for high-quality assembly.

Karen concluded that complete, telomere-to-telomere assemblies will revolutionise genetics and genomics and there is a lot of human biology “that requires a new look” with long reads.

Finding the needle: Haplotype-resolved annotation of clinically relevant genetic and epigenetic variants using Whole-Genome Sequencing – Sissel Juul, Philipp Rescheneder

Sissel and Philipp provided an overview of recent findings from the Oxford Nanopore Applications teams:

• A single Oxford Nanopore dataset enables the detection and phasing of genetic (SNPs, Indels, SVs, CNVs) as well as epigenetic variation (5mC, 5hmC) providing the future potential for precise diagnoses, better clinical management, and removing the stress of not knowing the cause of disease

• Haplotype-resolved variant calling enables the detection and accurate annotation of compound-heterozygous variants providing deeper insight into genomic instability in cancer

• Native long reads sequencing allows for accurate characterisation of repeat expansions on a single read basis as well as microsatellite marker regions

• Comprehensive variant analysis in cancer has been focusing on genome-wide methylation differences, and less than 5% of methylation differences would have been identified by arrays vs. whole-genome nanopore sequencing

• Highly accurate read-out of epigenetic modifications like 5mC and 5hmC enable a better understanding of imprinting diseases as well as tumourigenesis.

• The session closed with an overview of the team’s work on cell-free DNA analysis, including the application of short fragment mode, fragmentomic analyses, CNV calling, and identifying tissue of origin.

Cancer research breakout

Rapid leukemia classification using nanopore sequencing - Mélanie Sagniez

• Leukaemia is often fatal and has a lengthy diagnosis process - on average 14 days, with lots of different samples needed

• The subtypes of leukaemia are characterised by molecular rearrangements. Melanie highlighted acute lymphoblastic leukaemia (ALL), which is currently stratified using short read sequencing, to ask whether nanopore sequencing could improve ALL classification

• Melanie and her team used a combination of cDNA and RNA isolated from 12 ALL research samples and sequenced them using MinION and Flongle Flow Cells on the MinION

• Melanie concluded that the ALL subtype predictions “were really good” and highlighted they could predict ALL subtypes after 4 minutes of sequencing and suggested nanopore sequencing “has the potential to highly improve today’s standard” of ALL classification.

Shining a light on the FCGRs; deconvoluting their challenging (epi)genomic organisation and consequences in immune oncology research - Sarah Frampton

• Monoclonal antibody (mAb) therapy has revolutionised cancer treatment but not all patients are cured. Resistance and relapse are common. Fc gamma receptors (FCGRs) are important to the success of mAb immunotherapy

• Sarah highlighted very high sequence homology between FCGRs, so that short-read sequencing often fails to produce reads that can be adequately aligned or assembled. She highlighted how the cost and flexibility of the nanopore sequencing platform is a great benefit;they mainly used nanopore sequencing for the long reads and the ability to sequence native DNA as the methylation data “is a bonus’”

• Sarah also developed targeted sequencing approaches to study FCGRs: Cas9 and adaptive sampling to explore FCGR expression patterns between normal and malignant cells and the impact of FCGR variants

• Using nanopore sequencing, Sarah has been able to analyse SNPs which have been inaccessible with short-read sequencing. It was “quick and easy” and gave “an unprecedented view”.

Long read sequencing and variant prioritization of unexplained early onset colon cancer trios - Melissa Kramer

• Melissa and her team have been using nanopore sequencing to study SVs in colon cancer. SVs can alter methylation and gene expression, although the functional impact of most SVs remain unknown

• When there’s no family history of colon cancer, multi-gene test panels may not detect the drivers of colon cancer so Melissa and her team used the PromethION to sequence SVs which are “often missed by short read approaches”

• Melissa described how research samples were sequenced at ~30x coverage, >30kb reads were selected and a filtering workflow applied to streamline the identification of multiple variants

• Melissa also used nanopore sequencing for differential methylation analysis and highlighted how “phasing methylation provides allele specific context”.

Clinical microbiology breakout

Long-read sequencing assay allows accurate characterization of the HIV-1 reservoir - Laurens Lambrechts

• There are 38.4 million people currently living with HIV-1, and although it is a treatable disease, it is still not curable. Sufferers have a persistent HIV-1 reservoir in CD4 T cells; each cell carries a different HIV variant, and some will be defective. Laurens explained that if a sufferer stops taking treatment, the virus comes back in a matter of weeks which necessitates the need to understand this reservoir

• The previous workflow to characterise this viral reservoir relied upon nested PCR of 9 kb (92%) of the HIV-1 genome length, and short-read sequencing; this was costly and labour intensive, so they turned to nanopore sequencing

• They developed a new assay to enrich infected cells and to tag different viruses with UMIs. Nanopore sequencing of HIV-1 DNA (R10.3 flow cell) resulted in 99.99% accuracy at 15x coverage, allowed development of a scalable and high-throughput assay with no size bias, and benefitted from a 10-fold decrease in cost per genome.

Chan-Zuckerberg ID: an open-source cloud-based pipeline and analysis service for metagenomic pathogen detection and monitoring - Katrina Kalantar

• The CZ ID collaboration focuses on the detection, identification and tracking of infectious disease. The CZ ID software is an open-source, cloud-based metagenomics platform for researchers that provides a simple data upload and cloud-based data processing system, data management for samples and projects, single sample reports of microbial abundances and multi-sample analysis and flexible downloads for offline analysis

• The CZ ID software can accept DNA and RNA sequencing data from any sample type or host organism, and has been developed with an Oxford Nanopore Technologies-compatible pipeline for metagenomic analysis

• Katrina demonstrated that the new pipeline can accurately identify organisms for a variety of sample types and explained that the beta version of the pipeline will be available in early 2023.

Analysis of clinical research samples by nanopore sequencing with Q20 chemistry reveals inaccurate classifications within the genus Serratia - Levin Joe Klages

• Serratia marcescens is a facultative anaerobic, gram-negative, bacteria, implicated in hospital-acquired blood infections. This strain is famous for its blood red colour when cultured

• Levin explained that bacteria were usually identified through blood culture to isolate and phenotypically identify the bacteria followed by biochemical and MALDI MS analysis; this can take days. Using these methods, 10 samples were identified as S. marcescens, but were white in colour when cultured rather than the characteristic blood red

• Cultures of these samples were used to isolate DNA, and assemblies were generated with R.9 and R10 pores. Levin noted that the sequencing was performed in under two hours. Whole genome comparisons were performed and showed that the ‘white’ samples were within the Serratia genus, but were not Serratia marcescens. They found that their samples were more closely related to S. navei and even discovered a new species which they aptly name S. sarumanii, after the white wizard in The Lord of the Rings

• Levin concluded that Oxford Nanopore sequencing has the potential to identify pathogens rapidly, increase identification accuracy and identify new species.

Innovations in targeted sequencing breakout

Rapid genomic analysis for thrombotic microangiopathy using nanopore adaptive sampling with EPI2ME SeqOne framework - Laurent Mesnard

• Current methods for detecting thrombotic microangiopathies (TMA), based on short-read sequencing, have long turnaround times (TATs) of around three weeks. “Early detection can direct the patient to the correct therapeutic approach and prevent organ loss or patient death”

• Laurent explained the future potential of Oxford Nanopore sequencing to transform the clinical scene of TMA with its rapid TAT thanks to real-time nanopore sequencing with in-silico enrichment via adaptive sampling

• With in-silico targeting of around 300 genes, they could detect and resolve complex genetic aberrations in TMA in around 3 hours

Reduced-Representation Methylation Sequencing (RRMS): real-time targeted methylation sequencing with Oxford Nanopore - Rocio Esteban

• RRMS enables cost-effective, direct characterisation of methylation patterns across important regions in the genome

• Library preparation is simple and streamlined, no enrichment steps needed

• Adaptive sampling increases coverage of target regions during real-time sequencing

• Off target reads can be used as shallow whole- genome sequencing reads

• Compared with RRBS and EPIC, RRMS produces more even coverage and recovers more CpGs.

CRISPRoff as potential treatment strategy for collagen VI congenital muscular dystrophy - Franziska Haarich

• Collagen VI congenital muscular dystrophy is an early-onset disease that involves muscle weakness, joint laxity, and progressive contractures

• Genetic basis e.g., COL6A2 recessive and dominant-negative disease-causing variants

• Leveraging the benefits of long nanopore reads, Franziska called variants over a 58 kb region and phased the COL6A2 using cas9 targeting

• Franziska introduced a proof-of-concept method for treating COL VI-CMD using CRISPRoff for methylation induced allele specific knockdown of disease-causing variant, followed by Cas9 enrichment targeted nanopore sequencing.

Challenges in germline variant interpretation for the molecular diagnosis of disease - Katie Dixon

• Katie highlighted how the genetic architecture underlying cancer susceptibility is complex, and family history is a significant risk factor. She and her team assessed the potential clinical utility of long nanopore sequencing reads for hereditary cancer genomes

• Using PromethION Flow Cells, they characterised complex SVs and epigenetics variants and Katie highlighted how she and her team identified previously unknown variants which may be causative

• Nanopore sequencing has improved the detection and resolution of genetic and epigenetic variation in coding and non-coding regions of the genome

• Integrating nanopore sequencing reads with population-based phasing has enabled characterisation of point-of-origin pathogenic variants where there is no family history

Single molecule methylation profiles of cell-free DNA in cancer with nanopore sequencing - Billy Lau

Billy Lau discussed the role of characterising methylation in colorectal cancer samples from cell-free DNA (cfDNA). The conventional methods for screening/monitoring for colorectal cancer are colonoscopy and CT imaging but there are critical gaps in these technologies, the most crucial of which is a lack of molecular information on the cancer. Liquid biopsies are a non-invasive way to detect and analyse cancer but traditional methylation analysis introduces challenges such as DNA damage, GC skew from PCR and artifacts from bisulfite conversion. Billy’s team has developed a method for high-throughput sequencing of cfDNA with Oxford Nanopore sequencing, which overcomes these challenges, as it detects DNA methylation without amplification or conversion.

For assessing methylation in cfDNA, Billy used clinical research samples from 20 patients from the Stanford tissue bank and compared them to cancer-free cfDNA, using the Native Barcoding Kit (EXP-NBD196), Guppy v5.0.16 on High Accuracy Mode and Megalodon 2.4.0. He observed that there was a higher level of cfDNA reads in patients with cancer, and that cancer cfDNA has more hyper- and hypo- methylation than cancer-free cfDNA.

Billy expressed his future plans for longitudinal profiling of cfDNA from cancer patients, which he hopes will allow tracking of tumour burden in patients over time. His future plans also include large cohort studies and testing analysis of different fluid samples.

Oxford Nanopore technology update

As is customary the day finished with all the latest Oxford Nanopore technology updates. This was delivered by:

• Rosemary Dokos, VP Product Management

• Stuart Reid, VP Development

• Lakmal Jayasinghe, VP R&D Biologics

A full analysis of this year’s technology update and the recording can be found here.

There is still one full day left of our hybrid conference, so you can register for online attendance for the final day of live presentations here.