The future of AI: Machine learning working together with scientific expertise

In a new episode of the pharmaphorum podcast recorded live on site at WIRED Health in London in March, web editor Nicole Raleigh spoke with Elise de Reus, co-founder of Cradle, the generative AI platform that helps scientists to design and program proteins.

What if, instead of trial-and-error iteration, there was move to machine learning and a decoding language? It is a huge paradigm shift in the field of protein engineering.

A biologist by training, de Reus has always been fascinated by what microorganisms and biology can do in large human applications. Looking to overcome the high failure rate in the lab, Cradle was founded, together with Harmen van Rossum, to move away from the trial-and-error approach, and is spearheaded by a team that includes Google’s Stef van Grieken, Daniel Danciu, and Eli Bixby.

When it comes to challenges faced by scientists developing bio-based products today, GenAI can permit better protein design from the get-go, says de Reus, speeding up development over fourfold. These advances represent significant real-world gains when it comes to R&D of biological products, a process which has historically been extremely costly, also. Enzymes and proteins can offer sustainable solutions for planetary and one health, as well.

And when it comes to a future vision, Cradle wants this technology, state of the art machine learning, to be available to all – whether big pharma or start-up – and that these tiny but powerful machines be the best they can be in the process: machine learning working together with scientific expertise, not replacing scientists, but simply trying to give them better tools.

You can listen to episode 125a of the pharmaphorum podcast in the player below, download the episode to your computer, or find it - and subscribe to the rest of the series - in iTunes, Spotify, acast, Stitcher, and Podbean.