Brands
Latest top stories
Start-ups
Technology

Isomorphic Labs raises $5bn for AI drug design as first trials approach

18 November 2026

Image by Isomorphic Labs

 

Building on Nobel Prize-winning protein research and billion-dollar pharma deals, Isomorphic Labs is preparing to take its AI-designed treatments from the screen to the clinic. On 17 November 2026, it announced a $5bn Series C round to expand the platform and advance its drug programmes.

Greenoaks and Sequoia Capital co-led the round, with Andreessen Horowitz joining as a new investor and Alphabet investing again. The deal brings the total raised across the company’s three announced rounds to $7.7bn, following a $600m round in March 2025 and a $2.1bn round in May 2026. Isomorphic Labs declined to disclose its valuation, Reuters reported.

 

How AlphaFold led to Isomorphic Labs’ drug design engine

 

Founded by Sir Demis Hassabis within Alphabet in 2021, the company built on the breakthrough of DeepMind’s AlphaFold 2, which demonstrated that an AI model could predict a protein’s three-dimensional shape from its amino acid sequence alone. Hassabis and DeepMind researcher John Jumper went on to share half of the 2024 Nobel prize in chemistry for their work on protein structure prediction. By then, Isomorphic Labs and Google DeepMind had already unveiled AlphaFold 3 in the journal Nature, expanding the model’s reach to complex interactions involving proteins, DNA, RNA and small molecules. For drug researchers, this meant being able to map how a candidate medicine might lock on to a target protein.

Knowing where a molecule might fit, however, is only half the battle when deciding what to synthesize. The London-based company’s proprietary Isomorphic Labs Drug Design Engine (IsoDDE) goes a step further by estimating how tightly a compound will bind to its target, giving chemists a sharper filter for choosing which candidates merit costly laboratory testing. The same platform underpins work on both traditional small-molecule drugs and larger biologic treatments, such as antibodies.

In a technical report published in February 2026, Isomorphic set out where its new engine improves most markedly on AlphaFold 3. On the thorniest test cases (molecular structures least resembling anything in the model’s training data) IsoDDE accurately predicted the position of a small molecule roughly half the time, double AlphaFold 3’s success rate of about a quarter. That performance gap narrowed when the models were tested on more familiar structures. Separate benchmarks also showed that IsoDDE could uncover elusive "cryptic" binding sites; these are pockets that remain hidden when a protein is examined in isolation, opening up new avenues for drug discovery.

 

Pharma partnerships and the first clinical trials

 

Isomorphic Labs began forging ties with major pharmaceutical groups in 2024, signing small-molecule discovery deals with both Eli Lilly and Novartis. Lilly agreed to pay $45m upfront and Novartis $37.5m, with substantial further sums tied to research milestones. Novartis went on to expand its collaboration in February 2025, while Johnson & Johnson entered the fold in January 2026 with a multi-target research agreement spanning several types of medicine.

Alongside those commercial partnerships, Isomorphic is advancing its own internal programmes in oncology and immunology. The company says both have yielded encouraging preclinical results and that a portion of its new funding will go towards building the infrastructure needed to run clinical trials. Yet the funding announcement offered few specifics – disclosing neither the underlying data nor which candidate will be tested first nor when human dosing might begin.

As those candidates move toward the clinic, researchers will begin measuring how the compounds behave in the body and which doses patients can safely tolerate, laying the groundwork for wider therapeutic testing. If Isomorphic’s platform delivers on its promise, the payoff could extend far beyond a handful of successful treatments. By accurately predicting how drugs interact with the body’s machinery before a single compound is synthesised, the technology offers the prospect of compressing years of costly trial and error, opening up complex disease targets that were previously out of reach and accelerating the arrival of new medicines for patients.

 

 

 

Liked this article? You can support our independent journalism via our page on Buy Me a Coffee. It helps keep MoveTheNeedle.news focused on depth, not clicks.

👉 https://buymeacoffee.com/movetheneedle.news