Oxford Nanopore welcomed back first live audience in three years, as well as an additional 5,000 people online for day two of London Calling 2022

The first live day of this year’s London Calling conference saw over 250 people through the doors of Old Billingsgate and a further 5000 people registered to attend via the virtual platform. The first hybrid day of the conference saw speakers presenting across a mix of stages including Plenary, Breakout, Showcase Stage, Mini Theatre, Poster and Spotlight Sessions. Below is a summary of the highlights of the day.

Welcome

Gordon opened this year’s live conference by saying that in the three years since London Calling was last at Old Billingsgate, Oxford Nanopore has released over 200 products, increased the single-molecule raw read accuracy from 95% to 99.6%, and hugely improved scalability and output. This has been reflected in the productivity of the community, with over 1,000 publications in 2021, which is three times as many as in 2019. The Nanopore Community is working on projects ranging from high-throughput human genomics, to individual samples and answers that are fast enough to potentially be deployed in critical care on the same day.

Gordon also highlighted how nanopore sequencing is the only platform to enable native DNA and RNA sequencing for comprehensive real-time SNP, structural variant (SV), and methylation calling within a single dataset in fully scalable formats from pocket to population. Of course, COVID-19 had to be mentioned, with critical work in public health contributing to over 1 million SARS-CoV-2 genomes sequenced in 85 countries worldwide. The Nanopore Community made, and is still making, a significant contribution when the world needed it most. Gordon then officially opened the first day of live talks by banging the same gong used to celebrate Oxford Nanopore being listed on the London Stock Exchange in 2021.

The potential of ultra-rapid nanopore genome sequencing for critical care medicine

Euan Ashley described how he and his team of researchers from Stanford University School of Medicine developed a workflow for ultra-rapid nanopore sequencing that resulted in characterisation of pathogenic variants in under 8 hours. The work used Oxford Nanopore’s high-throughput sequencing device, PromethION 48, and Euan described how his team removed barcoding from the sequencing workflow and used the cloud to help with real-time basecalling and alignment, to reduce the time from sample to variant identification. Overall, Euan noted that he believes the study demonstrated that rapid identification of pathogenic variants has the future potential to guide physicians in clinical management, improve prognosis, and reduce treatment costs.

Identification of novel genomic structures and regulation patterns at HPV integration events in cervical cancer

Next, Vanessa Porter took to the stage and described how she used Oxford Nanopore sequencing to investigate how the integration of HPV affects structure and regulation in cervical cancer genomes. With nanopore sequencing, Vanessa and her team can assess the impact of HPV integrations on local methylome profiles. This is important because HPV integration is found in 70% of cervical cancer cases and can cause complex structural rearrangements that can be delineated and categorized using long nanopore sequencing reads. Vanessa explained how her research is showing that HPV primarily integrates close to oncogenes and can upregulate the expression of the oncogene. She went on to say how some types of HPV integration affect DNA methylation across regions of over 400 kb on the affected haplotype, which emphasises the importance of sequencing long DNA fragments.

Infectious disease breakout

Three fantastic speakers started the infectious disease breakout with their individual presentations, before a varying and broad panel discussion of their work.

Near-sample, real time metagenomics-based pathogen genome detection of life-threatening community-acquired meningitis-associated pathogens - Madjid Morsli

• Many different pathogens can cause meningitis, but bacterial meningitis is a deadly form, causing the biggest global burden. Madjid’s work is aimed at rapid identification of pathogens that may be associated with meningitis via genome sequencing using the MinION

• Depending on the pathogen, identification could potentially be achieved in under 2 hours, from extraction to result. Their results were then confirmed by standard laboratory tests, however, this takes two full days

• 70% of the pathogens present within 52 cerebrospinal fluid clinical research samples were identified within 20 minutes of real-time analysis

• They are working to replace their PCR protocol with their ‘one-shot’ genome detection, genotyping and antibiotic resistance profiling of community-acquired meningitis directly from cerebrospinal fluid clinical research samples.

Association of gut microbiome composition with severity of SARS-CoV-2 infection - Kalyani Amarendra Karandikar

• The gut microbiome and especially their metabolites, short-chain fatty acids, are important in modulating the immune response. Kalyani’s work investigated the range of bacteria present in the gut of COVID-19 positive and negative subjects. The bacterial genomes were isolated from faecal samples and sequenced using the MinION

• The team found that distinct microbiome signatures were seen within COVID-19 positive subjects

• They now aim to validate protective biomarkers using a larger sample set with the hope of designing appropriate therapies.

Potential clinical applications of adaptive sampling - Christian Brandt

• Bacterial vaginosis in pregnant women can lead to pregnancy loss or pre-term birth

• Bacterial indicators are depletion of certain Lactobacilli and presence of e.g. E. coli, P. aeruginosa, Group B Streptococci

• Current identification of bacterial vaginosis is culture-based which has ‘reporting delays’ and ‘low sensitivity and specificity’

• To detect bacterial vaginosis, Christian depleted human DNA in metagenomic samples, producing a ~1.7x increase in microbial DNA

• Christian also described an adaptive sampling method to enrich for small antibiotic resistance genes. In a 4-hour runtime, all carbapenemase genes were identified

• The main benefit of nanopore was real-time sequencing for a quick response which could potentially be essential in a hospital environment in future.

Cancer research breakout

At the same time as the infectious disease breakout, a fantastic line-up of some of the latest nanopore sequencing in cancer research was being discussed.

Nanopore sequencing shows potential for personalized oncogenomics - Kieran O’Neill

• Kieran’s team are investigating how nanopore sequencing with PromethION could be used to better understand the genomes and transcriptomes from research samples of individuals with advanced cancer, in line with their Personalised OncoGenomics (POG) project at the Genome Sciences Centre, BC Cancer, Vancouver

• The team are exploring how nanopore sequencing can be used to research epigenetic dysregulation in cancer, by performing methylation calling and phasing; Kieran explained the benefit of nanopore technology in providing methylation information ‘for free’

• In another example, Kieran showed how ‘you can really see what is going on’ with nanopore sequencing, when it comes to resolving human papillomavirus integration into the host genome (work from Kieran’s colleague Vanessa Porter)

• Lastly, Kieran explained how he has explored phasing allele-specific expression in RNA sequencing data using germline SVs derived from nanopore data (via the tool clair3), in a diffuse large B-cell lymphoma research sample

• In future, Kieran plans to expand this work to further demonstrate the potential capabilities of nanopore sequencing for personalised oncogenomics, and they ‘really hope’ that in the coming years nanopore sequencing could be a part of their routine POG workup.

Targeted nanopore sequencing ushers in the era of routine long-read sequencing in translational research laboratories - Abderaouf Hamza

• Abderaouf highlighted the benefits of adaptive sampling for cancer genomics research – targeted enrichment drastically reduces sequencing cost-per-sample compared to genome sequencing, allowing variant calling, methylation calling, variant phasing, and fusion detection, all in one assay

• His team performed adaptive sampling using the MinION Mk1B device, on a pan-cancer gene panel of targets comprising 360 genes + 10Kb of flanking regions (totalling 1.5% of the reference genome); they used the bioinformatic pipeline NanoAdapt for variant and methylation calling

• Abderaouf discussed how methylation data could potentially be used for tumour classification in brain cancers and he stated that adaptive sampling ‘is a valuable tool for SV detection’

• He concluded that the multi-modality of nanopore technology ‘makes it a really useful tool in cancer research’.

Sequencing tumoroids derived from subjects with colon cancer, including Oxford Nanopore WGS-strategy - Federica Di Maggio

• There were 1.9 million new cases of colorectal cancer and 935,173 deaths worldwide in 2020

• Federica explained how her team study tumoroids derived from colon cancer tissue, using whole-genome nanopore sequencing to investigate single nucleotide variants (SNVs), structural variants (SVs), and methylation profiles (with >20x depth)

• From the variant calling they have performed in 58 genes of interest, they found good correspondence with short-read data obtained from the samples – this is ‘a very good starting point’ – and they plan to extend their analyses to other genes.

A single platform for multi-omics - Dan Turner

Dan Turner, VP, Applications at Oxford Nanopore, opened his plenary presentation by highlighting the breadth of research being presented by the Applications team during London Calling, covering measuring telomere length to metagenomics. Unable to do justice to all of the studies in just one talk, Dan focused his presentation on a single cancer research study that encapsulates how nanopore technology can deliver novel biological insights through comprehensive multi-omic analysis. Dan’s detailed plenary will be available in full to watch on-demand next week.

Long-read transcriptome sequencing reveals isoform diversity across human neurodevelopment and aging – Rosemary Bamford

Neuropsychiatric disorders are a leading contributor to the global burden of disease, so Rosemary Bamford discussed how an integrated genomics approach is needed to further our understanding of neuropsychiatric disorders. This includes identifying genomic sequence variations, detecting methylation changes and gene regulation abnormalities, and transcriptome profiling. Rosemary is using brain tissue clinical research samples to investigate alternative splicing in the brain. Alternative splicing is an important contributor to transcriptional diversity, and ultimately influences brain development and the development of psychiatric disorders; compared to short-read sequencing, long nanopore reads allow entire transcripts to be characterised and so a better understanding of isoform expression in the human brain in health and disease. Rosemary has used nanopore transcriptome sequencing with PromethION to study isoform diversity across brain development and during aging; Rosemary said that Oxford Nanopore is a ‘game changer’ for full-length transcript sequencing.

Exploring the genomic and epigenomic landscape of acute myeloid leukemia with nanopore sequencing – Alberto Magi

Cancer is a complex multifactorial disease, characterised by wide ranging genomic and epigenomic mutations, which may change during the course of the disease and in response to therapeutic intervention. Alberto’s team used nanopore sequencing to fully characterise and understand the mechanisms behind clonal genomic and epigenomic evolution of acute myeloid leukaemias (AMLs) in three clinical research samples pairs, taken at time of medical identification and relapse.

AML is a blood cancer, which is responsible for over 150,000 deaths per year. The team performed whole-genome sequencing using five MinION Flow Cells per sample, delivering approximately 25-30x coverage. In addition, two novel computational tools were developed: GASOLINE for detection of somatic structural variants (SVs) from two samples; and PoreMeth for identification of differentially methylated regions by comparing nanopore methylation profiles for two samples.

Few relapse-specific SVs were detected between the sample pairs and none had a documented role in drug resistance. As such, Alberto moved to the methylation data, where he stated that “bisulfite-conversion methods limit the resolution of methylation analyses”. He revealed that traditional whole-genome bisulfite sequencing provided information on just 50-60% of CpGs, whereas using nanopore sequencing they could analyse approximately 99% of CpGs for all samples, with 90% covered by at least five reads.

Approximately 1,200-1,300 differentially methylated regions were identified for each sample pair – 30-40% of which were outside the resolution for detection by traditional whole genome bisulfite sequencing. Alberto concluded that nanopore whole-genome sequencing enabled simultaneous analysis of genomic and epigenomic alterations, and the identification of DMRs with “unprecedented accuracy”.

Human genome assembly and analysis using R10.4.1, Kit 14, and duplex data – Miten Jain

Miten described how, in 2021, thanks to chemistry, protocol, and throughput improvements in nanopore sequencing technology, and the work of research teams around the world, ultra-long read sequencing (defined here as reads above 100 kb) became routine. Ultra-long reads now permit the generation of high-contiguity de novo assemblies and UCSC, in collaboration with research institutes, will shortly release three new ultra-long-read Genome in a Bottle genome assemblies for community access.

Moving to the latest chemistry, Kit 14 with R10.4.1 Flow Cells, Miten presented data generated in the past few days, showing the generation of over 140 Gb of sequencing data in a single PromethION Flow Cell. This high output is achieved with lower DNA and library input requirements than for the previous chemistry.

The new chemistry also provided a notable improvement in raw read accuracy, and Miten described how this is particularly evident in homopolymer regions. Homopolymers lengths of up to approximately 20 bases were accurately called, and Miten pointed out that there are few homopolymer runs above that in the human genome.

The team is currently generating their first human genome assemblies using the latest nanopore chemistry and flow cells; however initial results using sheared genomic DNA, are reaching NG50 scores exceeding 20 Mb. They are also evaluating the use of duplex data and developing further variant calling and methylation analysis tools. With the developments presented here, Miten concluded that “phased, Q40+ accuracy de novo assemblies may be achievable”.

Oxford Nanopore technology update

As is customary the day finished with all the latest Oxford Nanopore technology updates. This was opened by CTO Clive Brown and this year he was also joined by:

• James Clarke, VP Platform Technology

• Lakmal Jayasinghe, VP R&D Biologics

• Stuart Reid, VP Development

• Rosemary Dokos, VP Product Management

A full analysis of this year’s technology update can be found here: https://nanoporetech.com/about-us/news/oxford-nanopore-delivers-technology-update-annual-london-calling-conference-bringing

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: https://nanoporetech.com/lc22