Yingkai Zhang

Introduction Click to copy section linkSection link copied! Kinases are an important class of proteins for their multifaceted roles in cellular function, from cell division, metabolism, immune response, signal transduction, etc. (1) The human protein kinase family consists of 518 members with distinct functional roles, and there is a wealth of biochemical and structural data available that we can leverage. (2) Kinases function in balance with phosphatases to regulate the signaling pathways.

Abstract Peptide-protein interactions are essential to biological processes and drug discovery, but selecting high-quality models from predicted complexes remains challenging due to high false positive rates (FPR). Here we introduce TopoDockQ, a topological deep learning model leveraging persistent combinatorial Laplacian (PCL) features to predict DockQ scores (p-DockQ) for accurately evaluating peptide-protein interface quality, aimed at enhancing precision and mitigating FPR in model selection.

Introduction Click to copy section linkSection link copied! Accurate binding affinity prediction to given protein–ligand complexes is of the greatest importance for the scoring function (SF) in the structure-based virtual screening (SBVS) method. The recent progress of machine learning (ML) has enabled great advancements in ML-based scoring function development.