AI-designed drugs by a DeepMind subsidiary are headed to human trials

Alpha Fold from Google DeepMind It has already revolutionized scientists’ understanding of proteins. Now, the platform’s ability to design safe and effective drugs is about to be tested.

Isomorphic Labs, a UK-based subsidiary of Google DeepMind, will soon begin human trials of drugs designed by the Nobel Prize-winning AI technology. “We’re getting ready to go to the clinic,” Isomorphic Labs president Max Jaderberg said April 16 at WIRED Health in London. “It will be a very exciting moment when we enter clinical trials and start seeing the effectiveness of these molecules.”

Jaderberg did not clarify the timeline, but it is later than the company had planned to begin human studies. Last year, CEO Demis Hassabis said the company would enter AI-designed drugs into clinical trials by the end of 2025.

Isomorphic Labs was founded in 2021 as a spin-off from Alphabet’s AI research company Google DeepMind. The company uses DeepMind’s AlphaFold, a leading AI platform that predicts protein structures, for drug discovery.

Proteins are made up of 20 different amino acids and are essential to all living organisms. Long chains of amino acids are linked together and folded to form the three-dimensional structure of the protein, which determines the function of the protein. Researchers have been trying to predict protein structures since the 1970s, but this has been a tedious process due to the very large number of possible shapes a protein chain can take.

That changed in 2020, when Hassabis and John Jumper of DeepMind presented stunning results from AlphaFold 2, which uses deep learning techniques. A year later, the company released an open source version of AlphaFold available to anyone.

In 2024, DeepMind and Isomorphic Labs released AlphaFold 3, which further advances scientists’ understanding of proteins. We have moved beyond modeling proteins in isolation to predicting other important molecules, such as DNA and RNA, and their interactions with proteins.

“This is exactly what you need for drug discovery: You need to know how a small molecule will bind to the drug, how strong it is, and also what else it might bind to,” Hassabis told WIRED at the time.

Since its launch, the AlphaFold platform has been able to predict the structure of almost all 200 million proteins known to researchers, and has been used by more than 2 million people from 190 countries. The breakthrough earned Hassabis and Jumper the 2024 Nobel Prize in Chemistry, with the Nobel committee noting that AlphaFold enabled a number of scientific applications, including better understanding antibiotic resistance and creating images of enzymes that can degrade plastics.

Earlier this year, Isomorphic Labs announced an even more powerful tool, which it calls IsoDDE, its drug design engine. In a technical paper, the company touts that the platform more than doubles the accuracy of the AlphaFold 3.

The startup has formed partnerships with Eli Lilly and Novartis to work together on AI drug discovery, and is also developing a “broad and exciting pipeline of new drugs” in oncology and immunology, Jaderberg said.

“The exciting thing about the molecules that we’re designing is that we have a much greater understanding of how these molecules work, and we’ve designed them to be very, very powerful,” Jaderberg told the WIRED Health audience. “You can take it at a much lower dose, and it will have fewer side effects, and off-target effects.”

Last year, Isomorphic named a chief medical officer and announced it had raised $600 million in a Series A funding round to prepare for clinical trials. At the same time, the company has built up a clinical development team. Its mission is to “solve all diseases.”

“It’s a crazy job,” Jaderberg said. “But we really mean it. We say it with a straight face, because we believe this should be possible.”