Pharmacophores from Macromolecular Complexes with LigandScout
Chemical-feature based pharmacophore models have been widely used in drug design over the last few years. Whereas the recognition of a set of chemical features from a set of bio-active ligands has evolved to a state-of-the-art method, the perception of 3D pharmacophores from macromolecular complex structures with bound ligands has hardly been recognized and has not been implemented as a fully automated procedure.
In this chapter, we present an approach that automatically derives 3D pharmacophores from macromolecular complex data using a well-defined set of chemical feature definitions. In a first step, small molecule ligands are chemically interpreted and extracted from structure data using improved known and new algorithms. Second, from the interactions of the interpreted ligands with relevant surrounding amino acids, pharmacophore models reflecting functional interactions such as hydrogen bonding, ionic transfer interactions or lipophilic contacts are created and projected into 3D space. Optionally, pharmacophore objects subsequently can be overlaid in order to find geometrically and chemically compatible pharmacophoric patterns occurring in several comparable complexes. The efficient overlay algorithm utilizes a maximum clique detection algorithm and an efficient, analytical alignment method. A sophisticated visualization allows the user to inspect the molecule and the pharmacophore, as well as to make changes. Finally, applications for inhibitor models of human rhinovirus serotype 16 coat protein and ABL tyrosin kinase are presented.
Pharmacophores from Macromolecular Complexes with LigandScout,
in Langer, Hoffmann (eds), Pharmacophores and Pharmacophore Searches, pp. 131–150, Wiley, 2006.