New Open-Source Package Makeshift Integrates NMR Data for Protein Dynamics Models
anshulkundaje · x · 2026-08-21
Researchers released Makeshift, a lightweight open-source Python package designed to lower the barrier for using Nuclear Magnetic Resonance (NMR) dynamics data in machine learning. The tool integrates data access, curation, and analysis, allowing users to retrieve and parse BMRB entries, perform chemical shift re-referencing, and predict secondary structure propensity. By re-implementing and validating several widely-used NMR calculations, Makeshift aims to pave the way for more NMR-based models like Dyna-1. A preprint is available on bioRxiv.
More from Research
- τ0-VLA: Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Search — sii-research · 2026-08-22
- ITSMBench Open Sourced: 50 Tasks for Multi-turn Agent Evaluation — Shahules786 · 2026-08-22
- Recursion releases Nesso: 10x faster binding affinity model for drug discovery — anshulkundaje · 2026-08-22
- Interview with Owain Evans explores emergent misalignment and LLM persona safety — sjgadler · 2026-08-22
- NVIDIA & Berkeley open source tactile robotics method T-Rex — DrJimFan · 2026-08-22
- Lightwheel and Hugging Face release 100k-hour human manipulation dataset — vanstriendaniel · 2026-08-22