Approximately 800 million people rely on cassava globally, either as a source of food or a source of income. But cassava is being devastated by two viruses, both transmitted by the whitefly: Cassava mosaic disease and Cassava brown streak disease. The Cassava Virus Action Project (www.cassavavirusactionproject.com) is a network of researchers, farmers and others, collaborating to use genomic technologies to improve the management of these cassava viruses. DNA analysis of the virus, quickly and close to the crop, can help farmers to decide what action to take. We have empowered local communities to take decisions that maximize their crops while also minimizing the spread of these whitefly-borne viruses. For the first time, farmers struggling with diseased cassava crops can take immediate, positive action to save their livelihoods based on information about the health of their plants generated using a portable, real-time DNA analysis device. The project aims to reduce the risk of community crop failure and help preserve livelihoods. Oxford Nanopore Technologies portable MinION was used to identify which strain of virus was destroying the cassava crops of farmers in Tanzania and Uganda as part of the Cassava Virus Action Project. As the MinION delivers the information in real time (compared to the usual three months), farmers were able to take action much faster. For example, one was advised to destroy the crop and plant a different variety that is more resistant to the virus. In this talk I will outline how our team is using supercomputing, genomics, mobile DNA sequencing technology and teamwork to impact the lives of millions. The team’s latest work to bring portable DNA sequencing to east African farmers has been featured on CNN, BBC World News, BBC Swahili, BBC Technology News, and the TED Fellows Ideas Blog.

Molecular identification of species has long been possible via marker gene sequencing but requires bulky instruments and controlled laboratory conditions which are impractical for environmental experiments in isolated places. Recent advances in sequencing technologies now permit rapid in situ sequencing, but development of specific protocol pipelines from DNA extraction to bio-informatic analyses is still required for efficient sample processing. Here, we chose a combination of 3 marker genes in order to identify 85 different coral colonies and their holobionte onboard the research vessel Tara. We developed specific protocols for DNA extraction and multi-PCR amplifications in order to process a large number of samples in a restricted space and time. We used Oxford Nanopore Technologies to sequence amplicons on a MinION device and developed bioinformatics pipelines to analyze nanopore reads on a simple laptop, obtaining results in less than 36 hours. Protocols and tools used in this work may be applicable at a larger scale for rapid environmental diagnosis of complex samples in the context of climate change.

The emergence of multi-drug resistant bacteria has made effective and targeted infection control in hospitals even more important to save the lives of at-risk patients. Hospital environments are known to be enriched for antibiotic resistant organisms, but their distribution and transmission across different sites and rooms remains largely untracked. As part of the MetaSUB consortium, we are using nanopore sequencing to study the resistomes associated with various hospital environments and develop baseline maps for hotspots and transmission patterns. By constructing near-complete genomes from metagenomic samples, we are able to identify novel plasmids containing resistance genes, as well as track their spread through hospital environments. This work is building toward the development of “Smart Hospitals” in Singapore where real-time surveillance of pathogens in the environment enables targeted and effective infection control measures.

Chromatin folding is increasingly recognized as a regulator of genomic processes such as gene activity. Chromosome conformation capture (3C) methods have been developed to unravel genome topology through the analysis of pair-wise chromatin contacts and have identified many genes and regulatory sequences that, in populations of cells, are engaged in multiple DNA interactions. However, pair-wise methods cannot discern whether contacts occur simultaneously or in competition on the individual chromosome. We present a novel 3C method, Multi-Contact 4C (MC-4C), that applies nanopore sequencing to study multi-way DNA conformations of tens of thousands of individual alleles for distinction between cooperative, random and competing interactions. MC-4C can uncover previously missed structures in sub-populations of cells. It reveals unanticipated cooperative clustering between regulatory chromatin loops, anchored by enhancers and gene promoters, and CTCF and cohesin-bound architectural loops.

A wealth of studies has shown that gene regulation occurs at different scales, ranging from the promoter to the localization of the gene inside the nuclear space. Our laboratory studies dosage compensation in the nematode C. elegans, in which a condensin-like complex modifies the three-dimensional folding and position of the X chromosome, and down-regulates expression of X-linked genes. To understand how chromosome folding impacts gene expression, we use chromosome conformation capture (3C) technologies. For 3C, chromatin is cross-linked before DNA is digested in situ by a restriction enzyme. The restriction fragments are then ligated together, leading to formation of “contact” molecules in which fragments distant on the linear genome are ligated together. The 3D proximity of fragments makes them more likely to get ligated. One of the main limitations of 3C is the fact that each end of restriction fragments can be ligated only once, meaning that one can only identify pair-wise interactions and multi-way interactions can only be inferred from contact frequencies averaged over many cells. To overcome this limitation and to characterize gene conformation in single cells, we use nanopore long-read sequencing, thereby capturing many contacts per sequenced molecule. Initial results using this technology will be presented.

The Galapagos Islands are situated in geologically young archipelago about 900 km west of continental Ecuador. Owing to their recent origin, their isolated location and strong conservation efforts, the islands boast a rich repertoire of endemic species of both plants and animals. Perhaps most famous among endemic species are the Darwin's finch, 18 species of passerine birds that evolved from a common ancestor in the last 1.5-2.0 million years in a remarkable adaptive radiation affecting body size and beak morphology. During his travels around the world, Charles Darwin visited the Galapagos and collected these finches that so beautifully illustrate his theory of evolution of phenotypic diversity due to natural selection. For a two week expedition in March/April 2018, we went to San Cristóbal island in the Galapagos to sequence birds representing all the major phylogenetic groups of Darwin’s finches. From each bird we collected a small amount of blood then isolated DNA, made sequencing libraries and sequenced birds representing seven species of Darwin's finches using GridION/MinION and 48 flow cells, for a combined yield of around 200 Gb. In this presentation I will present our de novo Darwin's finch genome assembly, inter-species contrasts for structural variants (SV), as well as our inference regarding the importance of such SV’s on the adaptive radiation.

G-quadruplexes (G4s) are RNA/DNA secondary structures, made from G-quartets. G-quartets are formed through a cyclic Hoogsteen hydrogen-bonding arrangement of four guanines with each other, and the planar G-quartets stack on top of one another forming four-stranded helical structures. Several researchers have confirmed their existence in the promoter region of several medically important genes, such as c-MYC and KRAS. Their role has also been shown to affect several post-transcriptional processes, including splicing. Previously we showed that our novel method FoLDeR (Footprinting Of Long Deazaguanine RNA), is able to identify two G4s that regulate the alternative splicing of the apoptotic regulator Bcl-X. However, the presence of 7-deazaguanine substitutions prevents splicing of RNA strands. To resolve this, incorporation of 7-deazaguanine must be restricted to functionally relevant regions. To achieve this aim, a method is required to detect 7-deazaguanine in RNA. By using direct RNA nanopore sequencing we were able to gather sequencing data of both normal and fully substituted 7-deazaguanine RNA. Initial results indicated that 7-deazaguanine causes a massive reduction in sequencing and basecalling rates. Further analysis via Tombo showed that this is due to detectable changes in the ionic current on 7-deaguanines, as well as surrounding nucleotides. This now allows us to programme the Tombo basecaller to relate the changes in currents to 7-deazaguanine. Further work will also be done on the 7-deazadeoxyguanine substitutions in DNA. Ultimately this will provide a new method to study the biological occurrence and function of G4s in splicing and transcriptional control in several disorders such as cancer and cardiovascular diseases.

Carbapenem-resistant Enterobacteriaceae (CRE) cause life-threatening infections that are difficult, if not impossible to treat. In order to understand carbapenem resistance and track its spread, we have collected and sequenced hundreds of CRE isolates dating back to 2007. In 2017, carbapenem-resistant Citrobacter freundii were isolated from four different patients at one Boston-area hospital. These isolates had a distinctive arrangement of transposons carrying the carbapenem-resistance gene KPC which was found to be unique to CRE from our study, with 11 out of the 12 isolates containing it having come from the same hospital (2008-2017). Using Unicycler and Pilon on a combination of Illumina and Oxford Nanopore sequencing from these isolates, we obtained finished quality genome assemblies, including fully closed and highly accurate representations of chromosomes and plasmids conferring carbapenem resistance. These assemblies are revealing highly dynamic interactions among two families of plasmids harboring this unique KPC-carrying structure, having moved among three distinct bacterial genera while persisting within a single hospital for a decade.