Eli Lilly is inviting early-stage biotechs to make use of drug discovery models that the pharma is providing via a new AI platform.
The machine learning platform, called Lilly TuneLab, allows selected companies to access drug discovery models that the Big Pharma said had been “trained on years of Lilly's research data.” In return for using the platform, biotech users will be expected to contribute “training data,” with the aim of “fuel[ing] continuous improvement for the benefit of others in the ecosystem and ultimately patients.”
To avoid either Lilly or its biotech partners giving away their prized proprietary data, the TuneLab platform—which is hosted by a third party—employs so-called federated learning. This means that companies are able to take advantage of Lilly’s AI models without data directly being shared by either side.
“Lilly TuneLab was created to be an equalizer so that smaller companies can access some of the same AI capabilities used every day by Lilly scientists,” Lilly Chief Scientific Officer Daniel Skovronsky, M.D., Ph.D., said in a Sept. 9 release.
Macrocycle therapy-focused Circle Pharma unveiled itself as one of the first partners for TuneLab, while metabolic-disease- and neuroscience-focused insitro said it was also involved in the program.
Insitro CEO Daphne Koller, Ph.D., said the AI-enabled drug discovery company was “excited to again partner with Lilly in bringing our machine learning capabilities to their unique dataset, so we can build best-in-class predictive models for small molecule properties, and bring the benefits of delivering better drugs faster to the patients who are waiting.”
TuneLab is the latest offering from Lilly’s Catalyze360 initiative, which was first unveiled by Lilly CEO David Ricks in January 2024. Catalyze360 also includes the pharma’s investment arm, Lilly Ventures, and its incubator arm, Gateway Labs.
“For many early-stage biotech companies, the promise of AI and machine learning in drug discovery remains just that—a promise,” Lilly Catalyze360 head Nisha Nanda, Ph.D., said in this morning's release.
“While the industry buzzes about the power of AI/ML to accelerate innovation, most small biotechs face a fundamental hurdle: they simply don't have access to the large-scale, high-quality data needed to impact decisions and train truly effective models,” Nanda added. “With Lilly TuneLab, we're not just sharing resources, we are also compressing decades of learning into instantly accessible intelligence.”
The pharma estimated that the total cost of the drug disposition, safety and preclinical data that has gone into the first iteration of TuneLab comes to more than $1 billion, which makes it “one of the industry's most valuable datasets used to train an AI system available to biotechnology companies.”
The longer-term plan for TuneLab is to further develop the platform's features in subsequent releases, with Lilly already trailing the addition of in vivo small molecule predictive models in the future.