The Bioinformatics Research Group of NII has used machine learning approach to develop a novel structure-based method (SG-ML-PLAP) for the calculation of binding energy between a small molecule ligand (potential drug candidate) and its receptor (drug target). Benchmarking on 3D structures of protein-ligand complexes with known binding affinities indicates that this new method significantly enhances the correlation between predicted and experimental binding energy values. SG-ML-PLAP is available as a user-friendly web server and it can predict strengths of drug-target interactions, even for new types of protein targets that it was not trained on. SG-ML-PLAP will be a valuable resource for AI/ML-guided structure-based drug discovery.
Reference:
Pal, S., Pal, A., & Mohanty, D. (2025). SG-ML-PLAP: A structure-guided machine learning-based scoring function for protein-ligand binding affinity prediction. Protein science: a publication of the Protein Society, 34(1), e5257. https://doi.org/10.1002/pro.5257