Docking-based virtual screening: Past, present, and future

Identifying molecular binders for protein targets through virtual screening is an active and rapidly expanding field, as the ligands discovered can serve both as molecular probes for mechanistic studies and as initial hits for drug discovery. Since pioneering work in the early 1990s, docking-based virtual screening (DBVS) has become a cornerstone of structure-based drug discovery and has achieved substantial success in identifying novel small-molecule modulators for diverse therapeutic targets. In this review, we first describe the major components of DBVS workflows, including ligand-binding site identification, chemical library preparation, and molecular docking methodologies. We then summarize recent advances aimed at improving DBVS performance, with a focus on template-based approaches, deep learning-based docking and scoring functions, and the emergence of large-scale and ultra-large-scale docking campaigns. Finally, we discuss current challenges and future opportunities for DBVS, outlining key directions for continued methodological innovation and for maximizing the practical impact of virtual screening in early-stage drug discovery.