Blog @ Illumina
Real scientists. Real commentary.

Live from the Asia-Pacific Scientific Summit 2014

Kahlil Lawless and Brett Kennedy
| May 01, 2014

Day one was a fantastic start to the Asia-Pacific Illumina Sequencing Summit here on the historic island of Penang. Guests from across the region and plenary speakers joining us from further afield are soaking up the atmosphere of the Eastern and Oriental Hotel which has stood on this spot since 1885, before humans even knew DNA existed.

Speakers today included a diverse mix of academic thought-leaders and members of the Illumina executive team. Their presence is a testament to the explosive growth in use of Illumina technologies across arrange of fields in the region, and a unique opportunity for our customers to express themselves at such a high level.

David Dimmock from the Medical College of Wisconsin kicked things off talking about their use of whole-genome and whole-exome sequencing. Dr. Dimmock has a number of publications that highlight their use of this technology including the well-known case of then- 15 month old Nic Volker, for whom an allogeneic hematopoietic progenitor cell transplant was performed, based on sequencing results. Nic has had no symptoms for four years – one of the most celebrated early breakthroughs for next-generation sequencing.

Almost five years on, they have made significant progress increasing the samples and reducing analysis times (now down to only 1-2 hours per exome for interpretation and final report). Significantly, they are seeing a 40% rate of what was described as “definitive diagnoses”. This rate reflects a dramatic increase in diagnoses this year in part attributed to moving to exome sequencing earlier in the process and a much improved knowledge of genes linked to phenotypes. We look forward to seeing how new HiSeq X Ten systems may improve this even further.

When it came to choosing between whole-genome (WGS) and whole-exome (WES) approaches, he compared the speed advantage of WGS with best times in their hands of ~36 hours in rapid run mode vs. ~4 days for WES. Timing can be critical, but WGS does come at a price premium. The key bottleneck affecting his team was results interpretation and reporting, so we look forward to some of the data sharing approaches they are working on to facilitate this process. If you are interested in learning from their experience, they have documented it in a Science Translational Medicine publication last year.

The iconoclastic Stephan Schuster enlightened us to his diverse projects in environmental and conservation studies. Early research from him and his team is revealing the possibility of using natural aggregators of microbial diversity and pathogens such as the common fly. Metagenomic approaches to sequencing the bacteria they carry on their carapace revealed the abundance and geography of pathogens in different environments, and the potential the mechanisms by which they move through our environments.

Other projects focused around the use of sequencing to investigate genetic diversity in different species, to infer evolutionary history, and implement this knowledge to overcome inbreeding depression when bringing endangered species back from the brink of extinction. Citing work on Tasmanian devils, Madagascan critters, and California condors, his research is having a direct impact on conservation efforts. Most impressive is the condor effort, a project for which a paltry 16 condors have been revived to over 400 and counting, and have been re-introduced into the wild. Through whole-genome sequencing, they were able to choose specific mate pairs to target maximal genetic heterogeneity and then do an F1 follow-up to confirm the richness of diversity. These studies even resulted in previously undemonstrated proof of avian parthenogenesis and what they have coined the ‘supercondor’ produced by this mechanism.

Sarah-Jane Dawson of the Peter MacCallum Cancer Centre then discussed her investigation of ‘liquid biopsies’, circulating free tumour DNA present in serum that may be used to investigate solid tumours in the body. To capitalize on the promise of genomic medicine, it is clear that it will not be practical to collect tissue biopsies to track tumour changes, particularly in response to therapies. An alternative method that monitors tumour burden without tissue biopsy that is reflective of inherent heterogeneity is an attractive proposition.

A very recent publication from Bettegowda et al. identified the amount of ctDNA present in a wide range of cancer types. Not unsurprisingly, late stage cancers were associated with greater amounts of ctDNA, but many early stage cancers could still be detected using this assay. Dr. Dawson has previously shown that using MiSeq, it is possible to identify ctDNA from plasma. In metastatic breast cancer, ctDNA levels show a greater dynamic range and correlation with changes in tumor burden, than did CA 15-3 or circulating tumor cells. Using ctDNA, detection could also predate imaging-based detection by up to 5 months, suggesting that tumor relapse and minimal residual disease monitoring could be significantly impacted by this innovative approach. Finally, in a 2013 Nature article they also showed good concordance of CNVs and mutant alleles between plasma and matched biopsy samples, and that it was possible to identify resistance mutations arising during treatment.

Jennifer Gardy from the British Columbia Centre for Disease Control used her well-polished media skills to wow the crowd with her investigations of Tuberculosis (TB) outbreaks. TB is not an historical disease as many think, Dr. Gardy estimates that 1/3 of the world’s population carry the organism and 10% of these will go on to have active TB (that really is 200 million people!). Given the connected nature of the world and ease of transmission, according to Jennifer, ‘TB anywhere is TB everywhere’ so the ability to monitor and manage this organism is a serious world health issue.

Using a combination of epidemiology and mathematics, she and her collaborators have developed tools that by incorporate phylogenetic trees produced by sequence data to infer transmission information. They can even back-calculate transmission events required to generate the observed network! They are now sequencing 20 years of isolates (about 5,000 samples) to explore the history of TB in British Columbia. This kind of study was infeasible until recently, as sequencing is significantly superior to historical variable number tandem repeat (VNTR)- based genotyping methods that do not accurately identify relationships to follow transmission. Using 48 samples/lane at 150x, or 768 genomes per HiSeq dual flow cell,  they have estimated costs to sequence the entire TB genome are just about equal to those needed to perform VNTR genotyping.  Dr. Gardy is now collaborating on using TB sequencing to simultaneously diagnose, speciate, assess epidemiology and drug resistance in one test. Promising data suggest sequencing may be able to deliver this comprehensive data set, a powerful tool in the fight against the health effects of this long-term human adversary.

We were also graced by the presence of Rob Elshire from AgResearch New Zealand. Dr. Elshire comes from a long and colourful history of agricultural genomics, and is one of the forefathers of Genotyping by Sequencing (GBS), also known as RadSeq. The presentation had more of a personal story and words-for-the-wise feel as he shared his personal journey and lessons he has learned along the way about how to innovate, optimise, and scale solutions in the agrigenomics world, where budget squeezes, complex uncharacterised species, and scarce tools are the norm. These sage words of advice from a sequencing veteran were not wasted on the crowd, many of whom are just beginning sequencing odysseys of their own. Rob openly advocated for free collaboration and demonstrated his dedication to making these methods available to everyone through open source journals, and placing all his hard-earned code up online for other researchers to use and improve upon. This really highlighted for me just how valuable the sense of community is amongst Illumina users, and how such sharing has been the cornerstone of driving the true value of what sequencing can bring, and indeed the very inspiration for having Illumina Scientific Summits.

When Professor Jun Yasuda presented his project and vision for colossal genomics data integration in Japan, people were so overwhelmed by the scope and impact of his undertaking that it was beyond comment. Ever since the Human Genome Meeting in Singapore 2013, I’ve personally looked forward to the distant day when electronic health records could be combined with large scale genomics and consensual data release to provide the first truly random and growing dataset which would form the basis for understanding benign variation, and enable association studies or unrivalled power. The Tohoku Medical Megabank Organization (ToMMo) makes that dream a reality, with a staggering 80,000 individuals to be interviewed, phenotyped, and genotyped, and a further 70,000 neonates to be tracked and studied from birth. The ongoing benefits to genomic health and patient care will be ripple out from this project for generations, and we look forward to seeing the wave of findings from this dataset start to filter though, as they have now completed the first 1,000 whole genomes.