Protein-based “target space” and ligand-based “chemistry space” have been studied as individual entities. With the exponential growth of the Protein Data Bank (PDB) and databases like Relibase that provide additional information, it is now possible to represent, visualize and study these “protein-ligand interaction spaces” in novel ways. However, the wealth of information that is contained in these databases has been under-utilized. Recent publications suggest that we can use computational methods to mine the protein-ligand interaction space and enhance the information contained therein.
Mining Protein-Ligand Interaction Space covers the following topics: 1. Creation and maintenance of protein-ligand interaction databases. Characterization of interaction spaces using descriptors such as protein-ligand interaction fingerprints or other protein-ligand descriptors. 2. Characterization of protein-ligand subspaces (such kinase interactions, aspartyl protease interactions, PDEs) and development of user-friendly/interactive tools to mine this information. 3. Mining novel interaction types in design such as cation-pi interactions, halogen bonds and bridged-water interactions and incorporation of that information in design. 4. Comparative analysis of protein-ligand interactions and application of them to scaffold hopping or hybrid design to increase potency, novelty, improve selectivity.