In Silico Laboratory: New and Updated Tools for Protein-Centered Drug Discovery

Computational tools for protein-ligand binding predictions are essential in modern drug discovery. This chapter provides an updated overview of the ProBiS tools, which identify binding sites, predict ligand interactions, and analyze conserved water molecules. Using an efficient maximum clique algorithm, these tools enable large-scale mining of protein structure databases, including the Protein Data Bank (PDB) and AlphaFoldDB, for drug repurposing and functional annotation. Recent advances include the ProBiS-Dock database, which provides precomputed binding sites for proteome-wide ligand-binding prediction, and the ProBiS-Dock algorithm, a hybrid flexible docking method. Additionally, ProBiS-H₂O and ProBiS-H₂O MD identify conserved water clusters crucial for ligand binding, while the ProBiS-Fold web server predicts binding sites on AI-modeled protein structures. We highlight applications of these tools in therapeutic research. ProBiS-H₂O has identified conserved hydration sites in enzymes, ProBiS-H₂O MD has revealed allosteric water networks in kinases, and ProBiS-Dock has facilitated virtual screening for cancer drug targets. ProBiS-Fold has enhanced the annotation of AlphaFold2-modeled human proteome structures. By integrating these tools into computational workflows, researchers can accelerate the discovery of new therapeutics. This chapter provides protocols for their effective use in structural biology and drug discovery.