Future perspectives

The two computational approaches presented in this article, hydrophobicity maps and SEED, were developed to precisely determine hydrophobic pockets and to dock fragment libraries. The hydrophobicity maps will be useful for the characterization of binding sites for the 3D structures [82] of the large amount of sequences that are emerging from the many genome projects. The location of a binding or association site can be predicted from the clusters of most hydrophobic points on the surface, and the size and ligand type could be estimated from the area and/or the volume of the binding site cleft.

Drug design is a fully multidisciplinary research field. Methodologies and procedures from different scientific disciplines support and cross-fertilize each other. A typical example is the combinatorial strategy for fragment-based design which is common to SAR by NMR [83,84], the multiple solvent crystal structures method [85], and SEED-CCLD [2,15]. A better understanding of the physical principles of solvation is needed to help in designing drugs. Continuum models of solvation effects are particularly useful for docking; their use will grow significantly in the near future. The continuum electro static approach implemented in SEED allows to efficiently dock a library of molecular fragments to a receptor of known structure. The SEED-CCLD strategy uses combinatorial principles to construct candidate ligands. Possible applications are for de novo design, as documented here for thrombin, lead optimization, and the selection of monomers for parallel synthesis and combinatorial chemistry.

Although there is not yet a computational approach to step directly from genomes to drugs [86], we think that the methods and procedures described in this issue of Perspectives in Drug Discovery and Design are useful new developments for drug discovery.

0 0

Post a comment