Isomorphic Labs, a Google DeepMind spinoff, unveils IsoDDE, a system that it says surpasses AlphaFold 3 in biomolecular structure prediction and drug discovery
Today, we are excited to share an update on our progress towards a new frontier of drug design. We have unlocked a new paradigm …
IsoDDE is consequential chiefly because it shifts the claimed performance frontier from a broadly recognized predecessor to a product developed by the spinoff itself. The report does not establish downstream drug-development outcomes, so the claim remains a technical positioning statement pending broader validation.
First-order effects
Isomorphic Labs gains a new basis to position its drug-design platform against AlphaFold 3, while Google DeepMind’s earlier model becomes the explicit benchmark.
Drug-discovery partners and prospective customers have a new claimed capability to evaluate, but will need to assess IsoDDE’s performance for their own molecular-design workflows.
Second-order effects
Competing AI drug-discovery teams face added pressure to show that their structure-prediction systems translate into better design decisions, not only model benchmarks.
The move strengthens the commercial importance of proving performance on biomolecular interactions, where model outputs must be useful enough to influence experimental priorities.
Third-order effects
If newer proprietary systems repeatedly outperform foundational structure models, value may shift from general prediction breakthroughs toward integrated drug-design platforms that can turn predictions into actionable programs.
The sector’s differentiation will increasingly depend on independent validation and evidence of real drug-development utility, rather than on a single headline benchmark.
The trend: AI drug discovery is moving from landmark biological prediction models toward proprietary, end-to-end systems competing on whether they can improve molecular design decisions.
The drug design engine we're building at @IsomorphicLabs is extending the SOTA further across key benchmarks, showing huge progress in accuracy and capabilities critical for in-silico drug discovery. Incredible work from @maxjaderberg and the entire team at Isomorphic Labs!
The more I read up, the more impressive the breakthrough Isomorphic labs has made here. Isomorphic Labs' IsoDDE doesn't just predict protein structures better than AlphaFold 3, it can find hidden binding pockets in seconds that used to take six months of lab work, and predict how…
So proud of the @IsomorphicLabs team and incredibly privileged to have played a part in building IsoDDE. Huge milestone! 🚀 Check out what we've been working on below.
Among other things, IsoDDE's binding affinity predictions exceed gold standard physics-based methods—an important step toward AI-driven rational drug design. Grateful to have contributed to this work alongside the Iso team, and hugely excited for the future we're building
Founded by Sir Demis, Isomorphic Labs are continuing their work on AI drug design. From the technical report: 'Our results in structure prediction present the first step-change improvement since the release of AlphaFold 3. They enable structure-based drug design on the most [imag…
The Iso team has cooked something incredible: our new technical report unveils the latest results from our drug design engine, the IsoDDE, progressing far beyond AlphaFold 3. This breaks new ground compared to AF and other similar methods by a significant degree across all key [i…
Today we share a technical report demonstrating how our drug design engine achieves a step-change in accuracy for predicting biomolecular structures, more than doubling the performance of AlphaFold 3 on key benchmarks and unlocking rational drug design even for examples it has [i…
if you forgot, the CEO and founder of Isomorphic is Demis Hassabis. So ByteDance has just caught up (and slightly exceeded) AlphaFold 3, and the original team is... significantly ahead. [image]