AlphaFold2-Based Modeling and Docking Analysis of RORα-LBD and RORγ-LBD for Drug Discovery

The RORα-LBD (271-513) and RORγ-LBD (262-507) were modeled using AlphaFold2 based on the target proteins obtained from screening. The receptor grid was generated by sitemap, and the grid docking box was generated with default parameters. Energy minimization was performed for all 10 compounds using the OPLS3e force field. The LigPrep module was used to generate different tautomers, ionization states, loop conformations, and stereochemical features from each input structure, and the maximum number of conformations was set to 32. The candidate molecules were evaluated based on pharmacodynamics parameters using the QikProp module. Molecular weight, hydrogen-bonded donor, hydrogen-bonded ligand, solvent accessible surface area, K+ channel blocking related IC50, water solubility, lipid-water partition coefficient, oral bioavailability, and other ADME parameters were predicted and evaluated. The Lipinski 5R principle was used for compound filtration. High affinity ligands were selected based on gscore, and the complexes of the molecules and target proteins were obtained for further analysis.