Field notes from the 5th Eagle Genomics Symposium 2015

By Marek Kultys, 27 March 2015

Great expectations and serious opportunities ahead of us. The 5th annual Eagle Symposium 2015 took place on March 23rd at the European Bioinformatics Institute in Hinxton near Cambridge, and it was about so much more than just bioinformatics.

The main focus of the meeting was on bridging the gap between advances in bioinformatics research and the obvious benefits for human health in clinical or pharmaceutical setting. Speakers represented a whole spectrum of expertise, ranging from policy making to academic research, from clinical practice to commercial pharma industry. A healthy cross-section.

Microbial genomics in healthcare

Sobia Raza (Interim Programme Lead for Science at PHG Foundation) discussed the importance of microbial sequencing in healthcare, which might seem a bit less obvious than sequencing human genomes, but in fact can prove equally important to public healthcare. First of all, Sobia made a very strong distinction between time frames available for investigating genomes in a research and in a medical setting — the former being measured in weeks and days, whilst the latter in minutes and hours. Sobia also stressed the potential importance of microbial sequencing for the effective preservations of antibiotics and tackling antibiotic resistance. We all nodded to one another, recollecting our own research done for the Longitude Prize 2014.

Stories of visual debugging

Simon Andrews (Head of Bioinformatics at Babraham Institute) presented a collection of fascinating case studies in lab quality control, when visual evidence gave indication of sequencing going wrong. Catastrophic situations are usually very easy to spot on visual read outs from machines or activity monitors. Subtle mistakes are a bit more difficult, but a keen observer would notice anomalies or surprising features in their data, which would sound an alarm about batch contamination, air blockage in pipings, or even wrong identity of tested samples. But the most difficult cases are those subtle and chronic, which makes them almost invisible to human consideration.

Simon concluded that researchers should share data from failed experiments as much, as they like to share data from successful ones. We really liked Simon’s talk because of his visual approach to quality control in a laboratory, as well as debugging of scientific experiments.

Structure and unstructured patient records

Richard Dobson (Head of Bioinformatics NIHR at King’s College London) focused in his talk on medical records, their importance in genomic context and difficulty in making good use of them. Rich gathered 1/4 million patient records and found that for many diseases up to 90% of relevant and important medical information might be unusable for data mining purposes, because it is recorded by clinicians as free text narrative; only about 10% of patient data is structured and available for database search. This is, obviously, a problem not only in clinical practice, but also in genomic research, as only a fraction of patients’ phenotypic data can be processed for correlations with genomic variations.

Variation in knowledge about variation

Vincent Plagnol (Head of Computational Biology and Reader at UCL) presented a fascinating study done on London’s Jewish Ashkenazi population. Doing field research and telephone interviews, trawling through patient medical records, sequencing members of families affected by IBD (Inflammatory Bowel Disease) and analysing multigenerational family trees, Vincent and his colleagues found out that often variants are reported to be much more prevalent in populations than they actually turn out to be after closer investigation. This shows really how little we still now about the variation in human genomes.

Data virtualisation and data silos

Will Spooner (CSO at Eagle Genomics) introduces Eagle’s new service — Eagle Core. It’s premise is to make best use of hundreds of genomic databases, each siloed in its own storage space, using various standard and legacy technologies. Eagle created Eagle Core to help alleviate this unavoidable fragmentation. Through data virtualisation, Eagle Genomics enable institutions to make better sense and use of the data they already have. We also liked the clarity of the user interface of Eagle Core.

Keynote: more collaboration

Ewan Birney (Joint Associate Director of EMBL-EBI) gave an excellent keynote explaining why exactly institutions such as EBI are instrumental in making good use of genomic data. Because it is thus not just about storage space — it is also about interrogating this same data in a number of ways each time we want to learn something new about ourselves. There is no underlying theory that tells us how biological systems and structures will behave. Each time we want to understand a biological system we need to refer to raw data — there are no shortcuts. And because data is accumulating in an exponential fashion (doubling every 12 months), we really need to advance our abilities to make sense of it.

Ewan also called for further collaboration and integration of academic and commercial efforts. The CTTV centre has been established by EMBL-EBI, Wellcome Trust Sanger Institute and GSK to move some of the discovery of drug targets more towards the pre-competitive research environment.

If you would like to learn more about the events of the day, explore #egsym15 on Twitter. There is also a very useful archive of all presentation abstracts on the Eagle Genomics website.

The 5th Eagle Symposium was an inspiring event. We are looking forward to the 6th edition next year when — hopefully — we will be able to celebrate that so much more will have been done to bring us closer to the promise of genomic healthcare for everyone.

Looking for a good problem?

We are a close team of designers and researchers who are passionate about tackling ambitious and important problems. If you’re looking to grow your impact, we’d love to hear from you!

Get in touch

Privacy Policy