muni bio's autonomous AI agent generates three sub-nanomolar TREM2 binders

muni bio (Manifold Bio), partnering with Adaptyv Bio and powered by the NVIDIA Proteina-Complexa framework, has closed the loop on autonomous drug discovery with a self-driving research agent: the agent designs proteins on its own, and Adaptyv's automated wet lab validates them experimentally. Although the wet-lab stage takes days to weeks, the pipeline ultimately produced 3 sub-nanomolar binders against the target TREM2. What makes this notable is that it demonstrates a complete autonomous research pipeline—AI agent + computational design + automated wet lab—running end to end on a real task.

Confirmed

The agent performs protein design based on the NVIDIA Proteina-Complexa model and explored nearly 14,000 protein designs
In Adaptyv Bio's wet-lab validation, 9 out of 10 tested designs showed strong binding signals, with 3 reaching sub-nanomolar affinity
The 3 binders target TREM2 and outperform previous results on relevant leaderboards
The wet-lab stage took days to weeks to complete

Why it matters

This case shows how an agent can chain computational design and automated experiments into a closed loop, achieving end-to-end autonomous drug discovery rather than stopping at computational prediction
Multiple reposts emphasized that this workflow validates the feasibility of "autonomous AI research agents" in protein design, offering a template for automated closed-loop research on more targets

2026-08-19 ~ 2026-08-19 · 5 related posts

muni AI Agent Discovers Three Sub-Nanomolar Protein Binders — AllThingsApx · 2026-08-19
NVIDIA's Proteina-Complexa enables agents to discover sub-nanomolar protein binders — AllThingsApx · 2026-08-19
muni_bio agent explores 14,000 protein designs using NVIDIA tools — AllThingsApx · 2026-08-19
AI Agent Closes Design-to-Test Loop in Drug Discovery, Generating 3 Sub-nanomolar Binders — AllThingsApx · 2026-08-19
NVIDIA Autoresearch Agent Generates Sub-nanomolar Protein Binders — AllThingsApx · 2026-08-19