Model Releases
Run NVIDIA BioNeMo NIM Microservices for Protein Structure Prediction in Claude Science
NVIDIA’s BioNeMo Agent Toolkit integrates over a decade of life‑science models and libraries into agent-callable skills, and when combined with Claude Science and NVIDIA NIM microservices it allows AI
NVIDIA’s BioNeMo Agent Toolkit integrates over a decade of life‑science models and libraries into agent-callable skills, and when combined with Claude Science and NVIDIA NIM microservices it allows AI agents to run end-to-end protein structure prediction workflows that use multiple‑sequence alignment (MSA) and multi‑folding models. In benchmark tests on the Seh1 monomer and Seh1–C1HCX1 heteromer, both OpenFold3 and Boltz‑2 achieved high interface confidence (iPTM ≈ 0.8) when fed MSA data, but scores dropped to ~0.15 without it, underscoring the critical role of evolutionary alignment. Structural superposition further confirmed accurate placement of β‑strands in the WD40 context, matching experimental observations; users are encouraged to explore the toolkit on GitHub to extend these analyses to other complexes.
Related
- NVIDIA BioNeMo Agent Toolkit Brings Accelerated AI to Life Sciences Researchers in Claude Science
- Build an AI Scientist for Life Science Discovery with NVIDIA BioNeMo Agent Toolkit
- Add a Specialized Deep Research Skill to Agent Harnesses
Source: NVIDIA Developer | 2026-08-31