Machine Learning-Based Scoring Functions, Development and Applications with SAnDReS

Background: Analysis of atomic coordinates protein-ligand complexes can provide three-dimensional data to generate computational models evaluate binding affinity and thermodynamic state functions. Application machine learning techniques create assess potential energy affinity. These methods show superior predictive performance when compared with classical scoring functions available in docking programs. Objective: Our purpose here is review the development application program SAnDReS. We describe creation complexes. Methods: SAnDReS implements scikit-learn library. This for download at https://github.com/azevedolab/sandres. uses crystallographic structures, targeted Results: Recent applications drug targets such as Coagulation factor Xa, cyclin-dependent kinases HIV-1 protease were able predict inhibition these proteins. outperform Conclusion: Here, we reviewed through studies SAnDReS-developed programs AutoDock4, Molegro Virtual Docker AutoDock Vina.