If you’ve been following the targeted protein degradation space — and you should be, because it’s arguably the most consequential modality shift since antibodies — you’ve heard the pitch on molecular glues. Small molecules that recruit an E3 ligase to a disease target, forcing the cell’s own garbage disposal to destroy it. No need for a binding pocket. No need for traditional enzyme inhibition. Just… delete the protein.
The problem? Molecular glues have historically been discovered by accident. Thalidomide. Lenalidomide. Nobody designed those — researchers figured out what they were doing after the fact. Rational, prospective discovery of molecular glues remains one of the hardest problems in drug design. The interactions are ternary (three-way — drug, target, ligase) and the binding surfaces are shallow, transient, and notoriously difficult to model computationally.
Ternary Therapeutics — name on the nose — just raised a £3.5 million seed round led by daphni, with participation from Pace Ventures, the i&i Biotech Fund, and the UK Innovation & Science Seed Fund. The London-based startup is building an AI-driven platform specifically designed for the rational discovery of molecular glue degraders targeting previously undruggable proteins.
The Biotech Voyager
Early-stage biotech signals, personalized.
The signals that matter to you — contextualized and written directly to you — so you cut through the noise and immediately understand why it matters.
No specific drug candidates have been disclosed yet — this is seed-stage, platform-building territory. The therapeutic areas they’re focused on include immunology, fibrosis, and neurology, all of which are rich with targets that have resisted conventional small molecule approaches. The thesis is straightforward: if AI can model protein-protein interactions and predict ternary complex formation, you can move molecular glue discovery from serendipity to engineering.
The competitive landscape is heating up fast. Monte Rosa, Orionis, and Siduma are all chasing rational molecular glue design, and big pharma has been throwing enormous sums at the space — Novartis, Bristol Myers Squibb, and Roche have all made multi-billion-dollar moves in targeted degradation. But most of that activity has centered on PROTACs and bifunctional degraders. True molecular glue discovery at scale remains elusive, and any platform that can reliably generate novel glue scaffolds will have pharma BD teams lining up.
It’s early. Very early. But the intersection of AI and molecular glue design is one of the most intellectually compelling bets in biotech right now. If the AI can see what chemists can’t — those fleeting ternary interfaces — this becomes transformative. Worth watching closely as they move from platform to programs.
