Are claims of the success of AI in drug discovery backed by science? 

Optibrium, a developer of software and artificial intelligence (AI) solutions for drug discovery, recently announced the promotion of Dr Tamsin Mansley to President of its new subsidiary to lead its US-based team and business operations. DDW’s Megan Thomas spoke with Dr Mansley to learn more about what she considers the role of AI in drug discovery and the associated challenges. 

MT: What do you think the role of AI is in drug discovery – where are the exciting applications and where is it heading? 

TM: AI plays a vital role in the drug discovery process by enabling chemists to do their best work. For me, AI isn’t about taking responsibility away from the chemist, or taking the credit for designing a new blockbuster. Instead, its role is to support chemists to make better decisions by giving them access to the data they need. 

There are currently three key areas of chemistry to which AI can be applied: generative chemistry, synthesis chemistry, and the prediction of compound properties. These applications, respectively, address the following questions: What can I make? How can I make it? And is this compound going to be successful in my project?  

Many drug discovery organisations don’t yet have solutions to these questions within their workflows. For a more widespread adoption of AI, additional work is needed to back up AI-related claims with scientific rigour, separating hype from reality and driving forward AI innovation. 

MT: Have you observed any barriers to uptake? How can they be overcome?  

TM: For a new technology to achieve high levels of adoption, it needs to be as user-friendly as possible. It also needs to integrate seamlessly with the user’s existing workflows. These are two aspects that we’re acutely aware of, and it’s why we embed these values – ease of use and ease of integration – into our products.  

To achieve this, we typically work with partners to apply our technology to specific problems that they have. This allows us to demonstrate the effectiveness of our software at addressing their individual needs, whilst also ensuring our technology delivers in terms of user friendliness and ease of integration.  

MT: What are other key challenges in this sector and how can they be overcome? 

Bold claims have been made around the success and efficacy of AI in drug discovery. The problem is, such claims aren’t always backed up by hard science. This understandably leads to scepticism within the industry around the value of AI and its place within the drug discovery field.   

There is also a sense of threat amongst some scientists who worry that their role may one day be outsourced to a machine. However, I’d say that the truly great medicinal chemists of the future will be the ones who embrace AI and machine learning (ML) in drug discovery. 

On the flip side, it’s worth recognising that whilst AI can already add significant value to drug discovery, it’s no panacea for every issue. We need to curb misconceptions around what AI can achieve and be more mindful of where it can be useful – and also of where its limitations lie. 

MT: When developing novel research technologies, what does Optibrium have to consider to ensure a successful product? 

TM: Drug discovery companies typically strive to design drugs that are either first-in-class or best-in-class. The same is also true for software companies, like us, who develop new products that either address an unmet need or improve upon the way things are currently done.

To ensure we develop successful products there are three key factors that we take into consideration. The first is that our product addresses a specific customer need, or pain point, experienced within drug discovery projects. The second step to success is designing the product to be as user-friendly and intuitive as possible. Thirdly, we make sure that each of our products is backed up by rigorous scientific evidence. In our view, you can’t create a successful product without addressing each of these key elements. 

MT: The company currently works with more than 170 organisations worldwide, including five of the top ten global pharmaceutical companies – to what do you owe this success? 

TM: By addressing key challenges faced by drug discovery scientists – and doing so in a highly visual way – our software helps make the data easy to work with, enabling scientists to identify strategies to drive projects forward. 

But it’s not only our products that we can attribute this continued success to, we’re also incredibly proud of our experienced team who offer industry-leading support to our customers. We’re pleased to regularly receive feedback from customers about how much they appreciate the knowledgeable support received from our team. 

Biography

Tamsin Mansley is Global Head of Application Science and President of US subsidiary, Optibrium. Mansley is an experienced drug discovery scientist, having worked as a medicinal chemist at Eli Lilly and UCB Research, with a passion for applying computational chemistry and predictive modelling approaches for compound selection and design.

Related Articles

Join FREE today and become a member
of Drug Discovery World

Membership includes:

  • Full access to the website including free and gated premium content in news, articles, business, regulatory, cancer research, intelligence and more.
  • Unlimited App access: current and archived digital issues of DDW magazine with search functionality, special in App only content and links to the latest industry news and information.
  • Weekly e-newsletter, a round-up of the most interesting and pertinent industry news and developments.
  • Whitepapers, eBooks and information from trusted third parties.
Join For Free