Highlights from Genome Science 2019

Two of Fios Genomics’ Operations team, our Head of Research and Development Dan Halligan and Bioinformatician Laura Bennett, recently attended the Genome Science 2019 conference in Edinburgh. This was a great opportunity to catch up on the latest developments and technologies used for genome science as well as highlighting some great applications of these for research. Notably, this year there was a strong emphasis on both long-read sequencing (with >2Mb now achievable with Nanopore) and single-cell genomics (and single-cell metagenomics).

Day 1

Day 1 of Genome Science kicked off with two keynote talks. First up was Wendy Bickmore (University of Edinburgh), who talked about the challenge of understanding the functional significance of genetic variation in the non-coding genome and the mechanisms by which enhancers regulate the expression of distant genes. The second keynote speaker was Sarah Teichmann (Wellcome Sanger Institute) who talked about the Human Cell Atlas project, a “periodic table of our cells”, and how single-cell genomics was enabling the identification of the full repertoire of cells and their co-location.

The final two sessions of day 1 focussed on Evolving Technologies. There were talks from Nanopore, Illumina, and BGI. Almost all discussed how quickly their technologies have evolved during the last 10-20 years. A perfect example of this was given by Vince Smith of Illumnia, who stated that their sequencing rate has progressed from ~1Mbp per day in 1998 to 1Gbp in ~30 seconds in 2017.

A dominant theme of the afternoon, and indeed the entire conference, was long read technologies. Mike Quail (Wellcome Sanger Institute) discussed use of PacBio for his research and how it has “come of age”. Matt Loose discussed the Long Read Club who aim to help the genomics community achieve end to end reference quality genomes routinely and at scale. One of the stand out talks of the afternoon was from Tuval Ben Yehezkel who described LoopSeq technology. LoopSeq attaches a barcode to each (long) DNA molecule, and then copies that barcode randomly within the same molecule. Once sequenced as short fragments, this enables the original long DNA molecule to be reassembled in silico. This effectively enables use of short read sequencing technology to generate long-read sequences.

Day 2

Day 2 started with a session on Evolutionary Genomics. The session featured a couple of talks with quite unusual topics. Alejandro Sanchez-Flores (National Autonomous University of Mexico (UNAM)), discussed a fish “tongue-eating” parasite, where the parasite actually replaces the host’s tongue. Secondly, Max Stammnitz (University of Cambridge), presented work on a transmissible cancer that causes facial tumours in Tasmanian Devils, which has wiped out around 80-90% of the population since the mid-2000s.

In the Clinical Genomics session Professor Sian Ellard (University of Exeter Medical School) discussed the 100,000 Genomes Project and the recent completion of the sequencing of 100,000 genomes from NHS patients affected by a rare disease or cancer. In particular, she highlighted NHS England’s new personalised medicine strategy with the ambition of genome sequencing as a routine service. There were two talks focussing on polygenic risk scores in coronary artery disease. Mike Inouye (Cambridge Cardiovascular) highlighted that a personalised polygenic risk score could motivate people to make beneficial changes to their lifestyle.

The Microbes session featured several talks tackling important issues in healthcare. In one of these, Tania Duarte (University of Queensland) described her research into the changes in the respiratory tract microbiome of Cystic Fibrosis patients. She is using long-read Nanopore sequencing to identify the effect of antibiotics on the microbiotic composition of the airway and importantly develop strategies for using sequencing as a tool for informing CF patient treatment. She highlighted a tool, Sketchy, for lineage calling and genotyping of pathogens from Nanopore reads.

Day 2 ended with a session on Bioinformatics and Software. Rachel Colquhoun (University of Edinburgh) presented a tool for graph-based variant calling, Pandora. Pandora allows variant calling across the pangenome of a bacterial species, therefore enabling the comparison of much more diverse sets of genomes. Anton Korobeynikov (St Petersburg State University) presented some new tools that have been recently added to the SPAdes genome assembler platform. SPAligner for aligning long reads to assembly graphs and PathRacer that aligns profile Hidden Markov Models (pHMMS) directly to the assembly graph. Apurava Narechania (American Museum of Natural History) presented an approach to horizontal gene transfer detection that avoids using alignment or trees but rather, adopts techniques from natural language processing and information theory. His proposed method is a model-free, unsupervised compression of k-mers into groups that can trace shared genome sequence.

Day 3

Day 3 focussed on Single Cell Genomics. Stephen Hague of 10X Genomics highlighted a recent improvement to their Chromium Single Cell Gene Expression Solution, Feature Barcoding technology, enabling the detection of more unique transcripts per cell. Martin Hemberg (Wellcome Sanger Institute) presented scfind, a search engine for genes in large single-cell sequencing collections. Zhouchun Shang (MGI/BGI) talked about DNA nanoball sequencing (DNBSeq), for dissecting cell heterogeneity.

The conference ended with two great keynote presentations. In particular, Jane Carlton (NYU) described her important work in tackling Malaria in India. While the social impact that taking the equipment and setting up labs and clinics in India was having was very impressive, a highlight was seeing the portability of the MinION sequencer in action. Jane showed a video of her post-doc on a Tuk Tuk in India with his laptop and the MinION as he sequenced a sample from a patient who potentially had Malaria, where time really was of the essence.