The covalent docking software landscape: features and applications in drug design

Covalent small-molecule ligands have re-emerged as powerful tools in drug discovery, offering prolonged target engagement, enhanced potency, and the ability to modulate proteins once considered undruggable. However, the rational design and virtual screening (VS) of covalent ligands remain challenging. Many docking tools cannot accurately model the energetics of covalent bond formation, often requiring more rigorous quantum mechanical (QM) or semi-empirical QM calculations for reliable predictions. Despite these limitations, the computational landscape is rapidly evolving. An increasing number of open-source, commercial, and web-based platforms now support binding mode exploration, lead optimization, and structure-based VS of covalent ligands. Alongside traditional approaches, new artificial intelligence (AI) and machine learning (ML) tools are assisting in prioritizing candidate molecules. This review introduces the fundamental principles and mechanisms of covalent inhibition, then provides a comprehensive overview of computational tools including covalent docking, warhead placement algorithms, and pharmacophore modeling, supporting early-stage drug discovery and chemical biology. Case studies highlight practical applications. We also cover curated databases of covalent binders and experimental 3D protein-ligand complexes, plus tools for assessing nucleophilic residue reactivity, all essential for robust covalent modeling. Finally, we briefly address risks associated with covalent chemistry. While progress is notable, further advances are needed. Nonetheless, today's covalent docking and AI-driven tools already make a meaningful impact by enabling rational design, generating new ideas, refining hypotheses, and expanding the boundaries of druggability.