Assessment of AI-based Protein Structure Prediction for the NLRP3 Target

Abstract

The recent success of AlphaFold and RoseTTAFold has demonstrated the values of AI methods in predicting highly accurate protein structures. Despite the advances, their roles in the context of small-molecule drug discovery need to be thoroughly explored. In this study, we evaluated whether the AI-based models can lead to reliable three-dimensional structures of protein-ligand complexes. The structure we predicted was NLRP3, a challenging protein target in terms of obtaining the three-dimensional model both experimentally and computationally. The conformation of the binding pockets generated by the AI models were carefully characterized and compared with experimental structures. Further molecular docking results indicated that AI-predicted protein structures combined with molecular dynamics simulations can be useful in small-molecule drug discovery.