‘An AlphaFold 4’ – Scientists are surprised by DeepMind drug spin-off’s special new AI
Isomorphic Lab’s proprietary drug-discovery model is a major advance, but scientists developing open-source tools are guessing how to achieve similar results
AI tools include predictions of how proteins interact with potential therapeutic molecules.
Google DeepMind releases an update after almost two years AlphaFold3 focuses on drug discoveryIts biopharmaceuticals spin-off, Isomorphic Labs, announced an even more powerful artificial-intelligence model — and they’re keeping it all to themselves.
Isomorphic Labs, based in London, touted the capabilities of its ‘drug-discovery engine’ – which it calls IsoDDE – in a 27-page article. Technical report, released on 10 February. Achievements, including accurate predictions of how proteins interact with potential drugs and antibody structures, have impressed scientists working in this field.
Yet unlike the AlphaFold AI system for predicting protein structure — which was made accessible to other researchers and described in depth in journal articles — ISODDE is proprietary, and the technical paper provides little information about how to achieve similar results.
On supporting science journalism
If you enjoyed this article, consider supporting our award-winning journalism Subscribing By purchasing a subscription, you are helping ensure a future of impactful stories about the discoveries and ideas shaping our world today.
“This is a huge advance on the scale of AlphaFold4,” says Mohammed Alquraishi, a computational biologist at Columbia University in New York City who is working to develop a fully open-source version of AlphaFold, referring to Google DeepMind’s unreleased future generation of the technology. “The problem, of course, is that we don’t know anything about the details.”
drug-protein interactions
AlphaFold 3 was developed with drug discovery in mind. on the contrary Nobel-prize winning predecessor AlphaFold2The model can predict the structures of proteins interacting with other molecules, including potential drugs.
Similar AI is modeled after AlphaFold 3 They have come closer to matching their performance and have acquired new abilities. An open-source model called Boltz-2, developed by scientists at the Massachusetts Institute of Technology in Cambridge and released last year, can predict how much strength potential drugs exert on proteins, or binding affinity. This is an important property for therapeutic development and is usually predicted by computationally intensive physics-based methods.
Isomorphic reports that its new AI outperforms both Boltz-2 and physics-based methods in determining binding affinity. Predictions of how antibodies – which form the basis of therapies selling tens of billions of pounds a year – interact with their targets are also state-of-the-art, the report claims.
Alqureishi says he is particularly impressed by ISODDE’s ability to predict drug-protein interactions of molecules that are significantly different from the data on which the model was trained. “This is a really difficult problem and it shows that they must have done something new,” he says.
secret sauce
Max Jederberg, president of Isomorphic, says the models behind IsoDDE are “profoundly different” from other efforts. But the company has no plans to reveal the ‘secret sauce’ behind it. “Like most big machine-learning and AI advances, it’s a combination of compute, data (and) algorithms,” says Jederberg. He hopes his team’s report will “inspire” the efforts of other teams building drug-discovery AI.
“This report comes after extensive efforts to partner with industry and potentially access their private structural data, so we do not know how impactful the additional data is on ISODDE’s performance,” Diego del Alamo, a computational structural biologist at Takeda Pharmaceuticals who is based in Cambridge, wrote on the social-media site X.
Isomorphic has signed drug-development deals with pharmaceutical companies Johnson & Johnson, Eli Lilly and Novartis potentially worth billions of pounds. It also has its own internal pipeline, with clinical trials in the near future. Jederberg says the company has developed different versions of IsoDDE used for technical reports, including working with its partners, to incorporate different data sources.
His colleague, Isomorphic’s director of machine learning, Michael Schaarschmidt, says the company’s data strategy is “quite broad,” including publicly available data, synthetic training data, and data sources that they “will try to license.”
Gabriel Corso, a machine-learning scientist who co-developed Boltz-2 and now leads the nonprofit Boltz in London, doesn’t think proprietary data played a necessary role in the reported performance of Isomorphic’s tool, based on his team’s benefits. “We can make a lot of improvements with the data that is available,” he says. “I think this is a new baseline to match – but also to pass.”
This article is reproduced with permission and was first published On 19th February 2026.
It’s time to stand up for science
If you enjoyed this article, I would like to ask for your support. scientific American He has served as an advocate for science and industry for 180 years, and right now may be the most important moment in that two-century history.
i have been one scientific American I’ve been a member since I was 12, and it’s helped shape the way I see the world. Science Always educates and delights me, and inspires a sense of awe for our vast, beautiful universe. I hope it does the same for you.
if you agree scientific AmericanYou help ensure that our coverage focuses on meaningful research and discovery; We have the resources to report on decisions that put laboratories across America at risk; And that we support both emerging and working scientists at a time when the value of science is too often recognised.
In return, you get the news you need, Captivating podcasts, great infographics, Don’t miss the newsletter, be sure to watch the video, Challenging games, and the best writing and reporting from the world of science. you can even Gift a membership to someone.
There has never been a more important time for us to stand up and show why science matters. I hope you will support us in that mission.
