Don’t downplay generative AI’s promise for drug discovery

Generative AI offers much greater potential than as a ‘point’ solution to a single problem, says Daniel Jamieson, CEO & founder of Biorelate.

Its broader appeal is as a smart decision support tool, joining up the R&D organisation and helping to hone drug selection.

The chance to employ deep-learning algorithms to distil complex knowledge into easily digestible summaries is understood in many industries now, but less so in life sciences.

This is a shame given the immense potential of generative AI (GenAI) to more accurately target drug discovery and its outcomes.

Today, the cost of bringing a novel drug to market can run into several billions.
Selecting the strongest candidates for commercialisation remains an expensive and protracted challenge that GenAI could help with, especially as a strategic R&D super-decision intelligence tool.

No drug developer can afford to waste time and resources repeating studies that have already been conducted, or pursuing problematic next-generation medicines – unless they can pinpoint the insights that will make a critical difference to success.

GenAI is a subset of artificial intelligence that, through deep learning, can very rapidly distil key information and insights from extensive and diverse knowledge banks, and create new output in a highly user-friendly format.

As a strategic part of drug discovery, the technology makes it possible to harness smart, mass-scale data analytics in a reliable and accessible way.

The opportunity here is to calculate which new drugs are worth pursuing, based on everything important that is already known about that field.

The issue up to now has been how to get to those insights reliably and efficiently.  Published research and other texts may contain rich knowledge, but for the most part this remains un-curated and impenetrable in scale.

Although cause-and-effect-like relationships may be abundant in the available text, for instance, they remain difficult to locate, connect and analyse.

In one scientific study, researchers may have examined how a specific drug triggers the activation of a particular protein.

Concurrently, another separate study may highlight that this activated protein is linked to the onset of hypertension.

While individually these findings provide valuable insights, it is only when they are connected that a potential hypothesis emerges: in this case, that the drug in question might pose a risk for inducing hypertension.

With GenAI, it’s possible to structure and analyse valuable data that up to now has been buried deep in internal archives as well as many millions of scientific research articles.

That’s in addition to all of the many other structured data sources, such as transcriptomics and proteomics.

Anything that can help drug development companies fail faster and improve the speed to approval and market access, by reducing the need for additional experiments, helps towards their goal to optimise spending and control costs along the development cycle, and get critical drugs to market faster.

At a macro scale across potentially hundreds of drug programmes, even just a 1-2% improvement in a drug’s chances of success, accelerated market delivery, or the chance to capitalise on additional opportunities, can have a huge impact across an entire drug discovery pipeline.

Targeted, appropriate application of AI to elicit valuable insights from across vast research archives can yield 10- to 100-fold improvements at certain parts of the clinical trials process.

Expediting early target selection to phase I clinical trials from what might have been four and a half years to just one represents a significant improvement in speed.

To date it has been rare to see a single business-level objective that transcends R&D and extends through to commercial planning in a way that can influence everything in an integrated way.

But this is likely to change, as companies come to appreciate GenAI’s fuller potential – as a kind of ‘chatbot’ or knowledge assistant for the entire organisation, providing an informed steer based on the big picture.

GSK has made good progress here, via its own GenAI large language models (LLMs) for specialist tasks, and a conversational interface that allows users to explore complex research questions without having to understand the company’s data ecosystem.

An LLM is a type of AI algorithm that uses deep learning techniques and vast data sets to understand, summarise, generate and predict new content.

GenAI offers a new way to interface with data and other AI models.

It allows users to ask – not only about the financial performance of biopharma companies and their drugs – but more probingly: “What are the best-performing drugs on the market, and what are their common mechanism of actions?”

AI continues to evolve, adding to the possibilities. In 2024, multi-modal algorithms (MLMs) will become a large and growing trend, presenting the opportunity for teams to interrogate not only text, but also images, sound and video.

More important than the potential applications though, is the robustness and reliability of the AI capability.

For GenAI to be trusted as a reliable ‘source of truth’ in drug discovery, knowledge sources – as well as the way that connections have been made - must be both transparent and validated.

Unless teams can absolutely trust the validity of the findings that are returned, and trace these back to their source, further painstaking assessments and analyses will always be needed.

It should be as much of a priority, then, to invest in the data’s integrity – and the ability to make supported claims about this – as to go all in with GenAI on the basis of its surface-level promise.

Daniel Jamieson is CEO of Biorelate. Go to biorelate.com

2024 PharmaTimes Clinical Researcher of the Year – The Americas focuses on AI. Excitingly, each submission will involve the creation of an AI-inspired model. To enter, search ‘PharmaTimes Clinical Researcher The Americas’