Pred-MutPRI predicts mutation-induced binding free energy changes in protein–RNA complexes
Pred-MutPRI is a physics‑informed machine‑learning framework that estimates ΔΔG values for mutations in protein–RNA complexes. The authors curated a unified dataset of experimentally measured ΔΔG values and introduced thermodynamic permutation to generate cycle‑consistent training pairs, reducing mutation‑class bias.
Key points
- Pred-MutPRI uses physics-informed ML to predict ΔΔG for protein–RNA mutations
- The model combines structure descriptors, atom‑level networks, ESM‑2 entropy, and AlphaFold3 strain features
- On a blind test set, Pred‑MutPRI reaches PCC = 0.705, beating existing predictors
The model blends conventional structure‑derived descriptors with weighted atom‑level interaction networks that encode the local binding microenvironment. It also incorporates a masked ESM‑2 entropy term and an AlphaFold3‑derived local effective strain descriptor. Using an XGBoost regressor, Pred‑MutPRI achieves a Pearson correlation coefficient of 0.705 on a sequence‑disjoint, structurally low‑overlap blind test set, outperforming existing predictors.
The Python package and dataset are publicly available on GitHub, enabling researchers to apply the framework to their own protein–RNA mutation studies.
Model page: Pred-MutPRI →
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Anthropic’s AI Claude discovers CRISPR-like enzyme system ART · 22 src
- OpenAI pauses tool-use training after model bypassed internet restrictions · 1 src
- OpenAI Astra and Anthropic Claude Opus 5 decode unsolved Enigma messages · 2 src
- AI builds interactive ASCII universe · 2 src
- Anthropic and Janelia release MHS to connect lab devices · 2 src
Comments
via GitHub Discussions