March 26th, 2019
Welcome to The Future Labs, a Podcast series talking to people building the future, today. We speak to experts around the globe about their role in changing the world for the better.
In this episode, we are exploring The Future of AI in Drug Discovery and how artificial intelligence is generating new ideas for new medicines. We are speaking to Jackie Hunter, Board Director at Benevolent AI a company that has developed a machine learning platform for drug discovery and development.
The development of new drugs and treatments is an extremely complicated and failure-prone process. The cost of bringing a new drug to market (including the cost of failures incurred along the way) is c.$2.7bn according to a 2014 estimate from the Tufts Centre for the Study of Drug Development. This high cost of development, and the decline in research and development productivity (termed “Eroom’s Law” – the inverse of “Moore’s law” of exponential growth in computing power) have resulted in many drug developers turning to external solutions to improve the efficiency of drug development.
Although computers have been a critical tool in drug discovery for decades, the past decade has seen a number of private companies take up the challenge of applying machine learning (ML) and artificial intelligence (AI) approaches to the problem. Benevolent AI is one of the best funded and well-known companies applying ML and AI to drug discovery and has built a platform specifically for this task.
In short, Benevolent AI’s platform is fed with large quantities of “structured” (i.e., curated) data and “unstructured” data from a wide range of sources such as scientific papers or proprietary datasets from pharmaceutical companies. The platform then processes this data to generate novel hypotheses about interesting drug targets and biology, to identify new applications for existing drugs (i.e., suggesting a different disease to treat), to design better clinical trials and ultimately to improve the probability of success of the drugs they are developing. The output of their platform is managed and interpreted by scientists who can then evaluate its quality and also conduct validation experiments in an actual laboratory.
Benevolent AI believes that their approach will not only make drug development quicker and cheaper, but that it may also generate ideas for new drugs that may not have arisen through traditional drug development approaches. Little is publically disclosed about Benevolent AI’s pipeline but Jackie discusses the pace of progress that the company has made since 2016 on its preclinical programs as well as the clinical stage drug in-licensed from Johnson & Johnson.
Listen to the Podcast to hear Jackie explain Benevolent AI’s approach to Drug Discovery and the progress they are making in bringing new medicines to patients.