NVIDIA's Proteina-Complexa enables agents to discover sub-nanomolar protein binders
AllThingsApx · x · 2026-08-19
NVIDIA's Proteina-Complexa framework, used within Manifold Bio's autonomous research pipeline, successfully produced sub-nanomolar binders for TREM2, topping the leaderboard.
Key Highlights of Proteina-Complexa:
- Unifies conditional generative modeling and optimization for atomistic protein binder design.
- No Redesign: Generates binder sequences end-to-end without inverse folding.
- Test-Time Scaling: Increased compute at inference leads to better binders.
- Wet Lab Validation: Screened >1M candidates for 133 targets with hit rates up to 63.5% and picomolar affinities.
Related event: muni bio's autonomous AI agent generates three sub-nanomolar TREM2 binders(5 posts)→
More from Embodied
- Unitree Founder Unveils 'Physical AI Robot Self-Evolution 1.0' at IPO Dinner — Scobleizer · 2026-08-19
- Human efficiency vs. robots: Walking a mile burns half a banana — noahsolomon · 2026-08-19
- Cybercab revolution imminent, launch event likely next week — mitchdeg · 2026-08-19
- Open Source Personal AI Computer: Local Compute Blueprint — tom_doerr · 2026-08-19
- Embodied-Navigator: Point, Think, Memorize, and Align for Navigation — OmniAI-ZJU · 2026-08-19
- Podcast: OpenRoboto on training humanoid robot AI on decentralized networks — markjeffrey · 2026-08-19