PocketSCP: A Method for Spatiotemporal Topological Visualization and Analysis of Protein Pocket Dynamics

The identification and analysis of pockets are crucial for understanding the functional mechanisms and therapeutic potential of proteins. However, it is challenging to track the dynamic characteristics of the pockets. In this paper, we present a method for the visualization and analysis of protein pocket dynamics called PocketSCP. Initially, the representation of lining amino acid atoms is proposed to characterize the spatiotemporal and topological properties of pockets. Subsequently, 3D mapping based on a reference molecular conformation is designed to generate 3D distribution maps of pockets. To facilitate observation and analysis, 3D to 2D plane mapping based on equidistant azimuthal projection is designed, leveraging the near-spherical shape properties of protein molecules. Finally, the efficacy of our method in identifying potential patterns within protein pockets is demonstrated through experimental validation.