Beyond borders 2017
Augmenting R&D with artificial intelligence
London-based BenevolentBio, a wholly owned subsidiary of BenevolentAI, is using artificial intelligence (AI) and machine learning to accelerate and improve drug discovery. Professor Jackie Hunter, a former SVP at GlaxoSmithKline, is BenevolentBio’s CEO.
EY: What impact could artificial intelligence and related machine learning tools have on the speed and cost of drug R&D?
Jackie Hunter: Artificial intelligence has the potential to impact the whole drug discovery and development process. As an industry, we're still losing 50% of compounds in Phase II and Phase III trials for lack of efficacy. That isn't sustainable; it tells us we're picking the wrong targets. A further quarter of failures in Phase II or III are for strategic or commercial reasons. That also tells us industry is not always making the right decisions about what compounds to prioritize.
Our deep-learning platform could lead to a fourfold increase in R&D success rates up to and including target validation. We already have some evidence for that: in less than a year, we have generated 36 new hypotheses and validated 24 of them in vitro. Traditional biopharma R&D would typically only manage about five in that timeframe with the same personnel.
We're also using our deep-learning supercomputer to generate chemistry models in less than a week, rather than a couple of months. It remains to be seen whether this acceleration translates to clinical proof of concept and beyond. But it's exciting.
EY: How does BenevolentBio's AI platform work, and what kinds of insights does it generate?
Hunter: The system ingests all kinds of scientific information - public, private, structured, unstructured - and annotates it with specialist biomedical dictionaries. Then we apply natural language processing and other algorithms to build a knowledge graph, showing the complex pattern of interactions between various molecular entities and diseases. This allows us to generate new potential associations or rule out existing hypotheses. Negative associations are sometimes even more valuable than positive ones, in terms of decisions to discontinue a particular approach.
The platform, a Judgement Augmented Cognition System, is trying to help us do more with what we know and to make better-informed inferences. It's not replacing the scientist or clinician, but rather enhancing and accelerating their hypothesis generation by helping extract relevant information from the vast mountains of data available.
We still need to test new potential associations in vivo, but the hope is that these have a greater chance of success and can thus dramatically speed up drug discovery. The idea is to generate fewer, better molecules whose properties we’ll be better able to predict, as well as better targets.
EY: That sounds like something most of big pharma would be interested in. Are you offering a drug discovery service?
Hunter: No. Unlike many AI companies working in the biopharma space, we’re not a service provider. We’re building our own pipeline. In November 2016 we licensed from Janssen a series of novel, clinical trial-ready small molecule candidates, along with a wealth of clinical and biological data. We’re using our platform to seek novel indications for these. The first will move into Phase II trials this year. Janssen has no buyback rights to these molecules, but they’ll get royalties and certain milestone payments if we move into Phase III.
In April 2017, we signed a two-year drug discovery collaboration with MRC Technology, a medical research charity. It will undertake complex chemistry on some of our AI-generated disease targets, and may also run promising molecules it has identified through our AI technology to validate. Previously, we licensed to a US pharmaceutical firm some targets and chemical scaffolds, generated using our platform, for use in Alzheimer’s disease.
EY: Investors have been piling into the broader AI space. What is your perception of the degree of investor and pharma interest in, and understanding of, AI as applied to drug R&D?
Hunter: AI is beginning to become more mainstream. We and other AI companies have raised significant venture capital. BERG Health [AI-backed drug R&D] is supported by Silicon Valley property billionaire Carl Berg. As for big pharma, most of them are dipping their toes into AI somewhere along the R&D value chain, whether in drug discovery, real‑world outcomes, or to better understand their customers. We are talking to a number of pharma companies about potential licensing deals around non-core assets.
EY: What is the biggest challenge you face in your quest to streamline and enhance drug R&D?
Hunter: The challenges are cultural and social, not just technological. Biologists must be open to the value that machine learning and data crunching can bring to their endeavor, and to asking new kinds of questions that may have previously been intractable. Data scientists need to talk to the biologists and chemists to better understand how their tools will be used. Big pharma needs to embed a more data-driven approach across all departments, not just within biostatistics or IT, to really benefit from what computing power and data analytics can bring to drug R&D.
- Pharma-digital DNA and why the term “digital health” will soon be obsolete by Lisa Suennen
- Streaming drug development — at scale by Matthew Gline, Roivant Sciences, Inc. – Coming soon
- Putting China’s capital to work in the West by Debra Yu, MD, China Renaissance – Coming soon
- Japan: leading the way in regenerative medicine by Yuzo Toda, Fujifilm Corporation – Coming soon