Isomorphic Inks Partners with Eli Lilly and Novartis for Innovative Drug Discovery Solutions

Isomorphic Labs, a London-based spinout from Google AI’s DeepMind, has announced strategic partnerships with pharmaceutical leaders Eli Lilly and Novartis to leverage artificial intelligence (AI) in the quest for new medications. This collaboration is valued at approximately $3 billion.

Under the agreement, Isomorphic will receive an upfront payment of $45 million from Eli Lilly, with the potential to earn up to $1.7 billion based on performance milestones, excluding royalties. Novartis has committed to $37.5 million upfront, along with funding specific research projects and performance-related incentives that could reach $1.2 billion over time.

“We are excited to embark on this partnership and utilize our innovative technology platform,” said Demis Hassabis, co-founder of DeepMind and CEO of Isomorphic, in a press release. “Our shared commitment to groundbreaking drug design and our appreciation for cutting-edge science make these collaborations particularly compelling.”

Fiona Marshall, president of biomedical research at Novartis, emphasized the impact of AI, stating, “Next-generation AI technologies have the potential to revolutionize drug discovery and expedite the delivery of transformative medicines for patients. This collaboration unites our strengths in AI, data science, medicinal chemistry, and comprehensive disease expertise to unlock new opportunities in AI-driven drug discovery.”

Founded in 2021 by Hassabis under Alphabet, Isomorphic Labs utilizes DeepMind’s AlphaFold 2 technology, which predicts protein structures in the human body. By revealing these structures, researchers aim to identify new pathways for developing treatments against diseases.

While the technology shows promise, a recent article in Nature Methods highlighted that AlphaFold can sometimes produce errors and functions more effectively as a “hypothesis generator” rather than a complete substitute for empirical data. However, its ability to generate accurate protein predictions at scale surpasses previous methods.

Researchers recently utilized AlphaFold to design a potential drug for hepatocellular carcinoma, the most prevalent form of liver cancer. Additionally, DeepMind has partnered with the Geneva-based Drugs for Neglected Diseases Initiative, a nonprofit pharmaceutical group, to apply AlphaFold in developing treatments for Chagas disease and Leishmaniasis, two significant health threats in the developing world.

In October, DeepMind announced that the latest version of AlphaFold can make predictions for nearly all molecules in the Protein Data Bank, the world's largest open-access biological molecule database. The model also accurately predicts the structures of ligands—molecules that interact with receptor proteins to alter cellular communication—as well as nucleic acids and post-translational modifications that occur after protein synthesis.

Isomorphic is already applying the new AlphaFold model, co-designed with DeepMind, to therapeutic drug design, focusing on characterizing various molecular structures pivotal for disease treatment. As the company moves forward, it faces pressure to achieve profitability; in 2021, Isomorphic reported a loss of £2.4 million (~$3 million) as it expanded its workforce in preparation for opening its new office in Lausanne, Switzerland.