Merck & Co. is taking a look at a generative AI platform. As an early user of Variational AI’s technology, the Big Pharma will assess the ability of the platform to generate novel small molecules that match its target product profiles (TPPs).
Variational’s Enki technology is a drug discovery spin on an idea that will be familiar to anyone who has used a generative AI platform. In a similar way to how AI software like DALL-E and Midjourney can create images based on text prompts, Enki generates small molecules in response to TPPs. The user picks targets they want to hit, as well as those they want to miss, and chooses other attributes. Enki then generates molecules that meet the TPP.
If Variational is right, only a series of prompts about the TPP stand between users and “novel, selective and synthesizable lead-like structures.” The startup trained Enki, using experimental data, to come up with molecules based on TPPs and therefore help researchers explore a wider swath of chemical space.
The history of startups making big claims about the promise of AI drug discovery platforms goes back years, and the hype has only intensified. Variational believes it is different, though.
“Adoption of AI for drug discovery is accelerating, but it is being led by companies using their proprietary AI to discover and develop their own assets,” Variational CEO Handol Kim explained in a statement.
“With the Enki Platform, chemists do not need to develop their own generative AI models, but can now submit their TPPs and get novel, diverse, selective and synthesizable lead-like structures in days to move quickly into lead optimization.”
Merck found the pitch sufficiently compelling to become an early-access user. The company, like its peers, is exploring the ways that teams across its organization could use AI. Robert Davis, the Big Pharma’s CEO, discussed how the company is thinking about AI and other technology at an investor event early this month.
“As you look longer term, how do we have to transform the business, and what capabilities do we need to build to do that? We’re making meaningful investments in artificial intelligence, machine learning, across what we’re doing in the labs, starting to think differently about how we approach customers,” Davis said.