Fios Genomics attended the Festival of Genomics in London this year which highlighted recent bioinformatics advances in the application of Next Generation Sequencing (NGS) data.
Since we have a diverse and skilled team of bioinformaticians in house keeping a close eye on developments in the field, it was great to see how much is already being performed but also how far we have to go to fully realise its potential. Presentations at this year’s conference covered topics from data sharing and enabling through to applications in the form of CRISPR and drug development.
Progress in the 100,000 Genomes Project
In the first session, Prof Sue Hill discussed how the NHS provides a great opportunity for the future of moving ‘omics technologies into the clinic. With detailed patient record data combined with a commitment to the 100,000 genomes project, now at >25k genomes, there is the promise of a great resource for the UK and healthcare as a whole.
It’s not hard to find concrete examples of how genomics data can help. In a presentation entitled “DNA differences: probably the most important medical advance” Sir John Burn noted that only 4,000 of 40,000 predicted cases of monogenic diabetes are diagnosed, and rates of diagnosis vary depending on location in the UK. While the 100,000 genomes project is ongoing, Chris Spencer’s (Genomics PLC) presentation emphasised how much can be achieved using data already available by performing a fascinating meta-analysis of 677 GWAS studies involving 2.7M participants, 14M variants and 545 distinct phenotypes in an attempt to generate a “wiring diagram” for humans.
Cost-effective NGS data generation
A second key theme was the substantial and ongoing reduction in costs of NGS data generation. Sir John Bell mentioned progress made by Oxford Nanopore and highlighted the recent advances in the technology (details are available in an online presentation). Similar strides are being made by Illumina with the release of NovaSeq.
However, cost reduction is not limited to these platforms but extends to diagnostic tools. For example, Sir John Bell also discussed aims to achieve a $20 WGS based diagnostic test for TB. With reduced costs of data generation, it’s clear that there are substantial challenges for bioinformatics. It was exciting therefore to see the explosion in platforms and computational tools for the industry to support bioinformatics analysis as well as companies providing analytical solutions.
Free access to data
Finally, in a session hosted by the Global Alliance for Genomics and Health (GA4GH), several presentations highlighted the need to standardise data access while incorporating end-to-end encryption. This is something that is vitally important when data is distributed across multiple data centres and hubs worldwide, each with potentially different restrictions on how data may be used and shared (Ewan Birney, now chair of GA4GH). Standardising methods of data access via web APIs can lead to increased interoperability of data stores, and examples for NGS data are being developed and implemented in NGS tools already (see htsspecs on github, covered in a talk by Thomas Keane).
Data visualisation
Finally, a discussion at the end of this session highlighted that representation of genetic variation through genome graphs has benefits today for improved variant calling but maybe some way from providing a full solution to the current reference based representation of genomes that’s so deeply ingrained in the file formats used for NGS data.
Dr. Daniel Halligan – Bioinformatics Team Leader at Fios Genomics
For an example of data visualisation available through Fios Genomics’ bioinformatics service, request a sample report.
