EOM 1352: Fri 7 June 2024, 11:36
Catalyst 7 July 2023, 14:52

Current Edition

Discovery Park: Wed 13 November 2024, 10:35
ramusmedical

Making AI Work for Drug Discovery: A Joined-up Approach

AI/data-driven drug discovery is starting to evolve at an encouraging rate now. But the technology’s positive impact will depend on how well the technology, and the insights it elicits, are embedded into R&D, says Biorelate’s Dr. Ben Sidders.

Gradually, data-driven, AI-enabled drug discovery is becoming a reality, beginning to fulfil the technology’s promise and demonstrating its potential more tangibly. The broader signs are encouraging, too – such as the growing profile of ‘AI-first’ companies such as Recursion and Insilico Medicine, and the observation that many traditional pharmaceutical companies are now embracing AI across their businesses.

Where previously the perceived value of AI in drug discovery and development had failed to live up to the technology’s hype, targeted solutions are now emerging which are making a positive impact on aspects of R&D. In turn, these applications are providing some valuable lessons and feedback about how to successfully embed AI within R&D operations.

In target discovery, knowledge graphs are now proving adept at integrating a vast number of data sources into a query-able structure, forming the basis for informed and relatively unbiased target prioritisation decisions and chemistry, where transformers are accelerating small molecule design and synthesis.

Challenges remain, however, predicting synergistic drug combinations has been the topic of extensive research, with only limited success and almost no translational relevance. Nor are we any nearer to being able to predict the effect of a drug on a given patient without first running a clinical trial.

The overriding realisation is that AI’s role in life sciences R&D is directly dependent on how decisively, and how well, they integrate the technology – and the insights it surfaces – within the wider R&D operation. Achieving this, in turn, will require a structured approach to AI-enabled R&D transformation, spanning four parallel priority areas: data, model, culture, and validation. Here’s how that breaks down.

Catalyst: Fri 8 November 2024, 14:16
Biosynth: Wed 13 November 2024, 10:18