Physics informed machine learning predicts binding free energy changes in protein-RNA complexes upon mutations

Researchers developed Pred-MutPRI, a physics-informed machine learning model that predicts changes in binding free energy (ΔΔG) caused by mutations in protein-RNA complexes. The model uses a curated dataset, thermodynamic permutation for balanced training, and integrates structural and sequence features including AlphaFold3 predictions. Pred-MutPRI shows strong generalization on unseen sequences, improving mutation impact prediction in gene regulation contexts.