SIUFocus: The Power of Big Data: Advancing Healthcare in the Electronic Era

Author: Stephanie Leonida Edited by: Inês Barreiros

The latest technological advancements in computational biology have led to an influx of raw, unfiltered data, accessible to researchers in large repositories or databases. Medical researchers and others look to these repositories endeavouring to identify specific drug targets and develop powerful gene therapies. Sifting through this pile of information requires time and money, which slows down drug testing in clinical trials and limits accessibility to innovative therapies. How can we accelerate this process and improve patient care?

On the April 27th, SIUFocus brought together three keynote speakers to address this question and share how they are exacting ways of revolutionising the healthcare system using artificial intelligence (AI) and data science techniques.

Our first speaker, Victor Dillard, COO of Desktop Genetics and registered expert on Biotechspert projected the necessity of marrying computer technology to biotechnology. At Desktop Genetics, Victor and his colleagues develop platforms to facilitate the development of CRISPR therapeutics. CRISPR-Cas9 is a unique genome editing tool that allows researchers to edit parts of the genome by removing, adding or altering sections of the DNA sequence. Although there was certainly no shortage in data generated through CRISPR technology, Victor realised what was lacking: a more efficient way of processing and using this genomic data to advance translational medicine. Through Desktop Genetics platform DESKGEN, customers in clinical research fields can design or get the company to design specific genome editing vectors. This allows researchers to build their own bio-toolkit for human genome editing of specific diseases. By pioneering AI – based technology, Victor and his team improve the efficiency of research and allow the rapid translation of CRISPR into personalised cell therapy. This is a remarkable example of how computer savvy regarding data science techniques can save time for scientists in optimising an array of genome editing vectors, which will ultimately favour safe and effective movement through clinical trials.

Our next speaker, Alex Gutteridge, Director in the GSK Computational Biology group, introduced yet more challenges facing innovations in healthcare. It was disconcerting and yet not surprising to discover that over 50% of Phase III clinical trials fail due to poor decisions being made before drug testing. Alex explained that much of these failures are due to lack of genetic validation supporting specific drug targets. One of the problems underlying this is ineffective translation of data accrued by Genome Wide Association Studies (GWAS). Again, we are faced with a large repository of unfiltered data. Alex and his team at GSK introduced an innovative integrated platform called Open Targets to tackle the problem. Open Targets is a phenomenal data mining tool that allows researchers to pull up evidence about the association of known or potential drug targets with diseases. Unique data filtration systems like this act as a driving force behind systematic drug target discovery and prioritisation.

Dr Parashkev Nachev (Senior Clinical Research Associate at the Institute of Neurology) finished the session with another revolutionary application of AI – based technology. Dr Nachev combines brain imaging data from UK Biobank with recorded biological behaviours associated with major illnesses like Parkinson’s disease, COPD, mood disorders, diabetes and more. UK Biobank’s recent initiative saw the release of imaging data from the world’s largest imaging study involving 5,000 participants. Brain imaging data can be accessed alongside measures of lifestyle, cognitive health, physical fitness and BMI. The crucial factor in outlining the success of Dr. Nashev’s research centres on being able to provide resolution between many clusters of people that share many disease traits. The aim is to produce signature brain maps to help identify and chart disease onset. In this way, using computational methods to build complex biological patterns can dramatically influence disease diagnosis and treatment to aid patient care. This is about harnessing predictive power of computer models in order to provide more rapid and effective pharmacological treatment.

These are just some of the ways computational biology is being used to make sense of data to inform important decision making in drug target discovery and therapeutic design. With these techniques, the current generation is likely to see a paradigm shift in medical research and medical practice that will forever change the healthcare system.