How artificial intelligence is the future of pharma
Posted: 5 December 2016 | Professor Jackie Hunter, BenevolentBio | 1 comment
Artificial Intelligence and machine learning present the industry with a real opportunity to do R&D differently, writes BenevolentBio’s Jackie Hunter…
There needs to be a fundamental shift in drug discovery and artificial Intelligence holds the key to bringing the pharma industry into the 21st Century.
The current drug discovery process needs to shift dramatically in order to meet the needs both of society and patients in the 21st Century. Artificial Intelligence and machine learning in particular, present the pharmaceutical industry with a real opportunity to do R&D differently, so that it can operate more efficiently and substantially improve success at the early stages of drug development.
The long term benefits of this will mean that the vast resources and money used to develop drugs in the current process will be deployed more effectively to give not only a better return on the investment but also a substantial increase in the delivery of new medicines for serious diseases.
The current drug discovery process – too lengthy and very expensive
It can take up to 15 years to translate a drug discovery idea from initial inception to a market ready product. This contrasts with the rapidity of innovation in other industry sectors. Identifying the right protein to manipulate in a disease, proving the concept, optimising the molecule for delivery to the patient, carrying out preclinical and clinical safety and efficacy testing are all essential, but ultimately the process takes far too long.
Industry is currently said to spend well over $1 billion per drug
Industry is currently said to spend well over $1 billion per drug. That’s partly because all the drugs that didn’t make it have to be paid for. Picking the protein target, developing assays to measure activity at the target and screening a large number of molecules to get the right molecule for the effect you want can take anywhere between two to five years.
This is before you can test safely in animals and then in Phase 1 testing on human volunteers. Importantly, even when the compound has got this far, the chances of it making it all the way through to the market are less than 1 in 10, even with years of research already invested.
“this lack of success is why so many companies have had to merge”
In short, the odds are not good. In fact, this lack of success is why so many companies have had to merge because, over time, the current drug discovery process is becoming less and less sustainable as a business model.
The role of AI and deep learning in the drug discovery process
The drug discovery process and the researchers that drive the pipelines can be greatly aided by the latest innovations in AI and machine learning technology. The average biomedical researcher is dealing with a huge amount of new information every day. It’s estimated that the bioscience industry is getting 10,000 new publications uploaded on a daily basis – from across the globe and among a huge variety of biomedical databases and journals.
So it’s impossible for researchers to know, let alone process, all of the scientific knowledge out there relating to their area of investigation. What’s more, without the ability to correlate, assimilate and connect all this data, it’s impossible for new usable knowledge – which can be used to develop new drug hypotheses – to be created.
AI and machine learning
AI and machine learning have a vital role to play in augmenting the work of drug development researchers so that an informed, first analysis of the mass of scientific data can be conducted in order to form essential new knowledge.
As a practical example, my own company BenevolentBio, has been doing research into Amyotrophic Lateral Sclerosis (ALS). The AI we’ve developed – embodied in the company’s Judgement Correlation System (JACS) – is able to review billions of sentences and paragraphs from millions of scientific research papers and abstracts.
JACS then begins to link direct relationships between the data and regulates the data into ‘known facts’. These known facts are curated, and hitherto unrealised connections made, to generate a large number of possible hypotheses using criteria set by the scientist – there were around 200 for ALS.
An expert team of researchers then assess the validity of these hypotheses and arrives at a prioritised list of hypotheses which are considered to be worth exploring. Further interrogation by the scientists whittles this down to 5 hypotheses that we then test in the lab – and some potential new mechanisms for disease modification are identified.
Accelerating drug discovery with technology
In terms of compound design, the scope and augmentation that AI and machine learning give us will mean that we can tap into a much broader chemical space, in turn giving us a much wider and more varied chemical palette to better enable us to pick the best molecules for drug discovery.
The technology will also help in terms of the industry’s selection of patients for clinical trials and enable companies to identify any issues with compounds much earlier when it comes to efficacy and safety. So the industry has much to gain by adopting AI and machine learning approaches. It can be used to good effect to build a strong, sustainable pipeline of new medicines.
About the author
Professor Jackie Hunter has held senior positions at global pharmaceutical organisations including GSK, Proximagen and OI Pharma Partners and joined BenevolentAI as CEO of BenevolentBio in 2016. Jackie has vast academic and business experience in the biomedical and pharmaceutical sectors. She directs the application of BenevolentAI’s technology for drug development and gives the company the insight it needs to operate its unique business model – one which sees it not only researching, but also developing the blueprint for new drugs.