Monday, November 20, 2000
Over the last five years, structure-based drug design has fallen
out of favor because protein structure determination was too slow and expensive.
In its place, high throughput screening has come into vogue as a drug discovery
tool in the belief that it would be superior to structure-based approaches.
But new structural genomics initiatives promise to bring structure-based design
back to the forefront with the structures of large numbers of new targets available
either in the public domain or for sale.
Structure determination has been slow, taking two years and
more than $100,000 per structure. Besides the time and cost, the available structures
have been limited to the few proteins that have been amenable to either X-ray
diffraction or nuclear magnetic resonance (NMR) analysis. Indeed, the Protein
Data Bank (PDB), a public archive of biomolecular structure information supported
by the National Science Foundation, the Department of Energy and the NIH, contains
about 340 unique protein folds or topologies represented by about 12,000 different
proteins, minuscule in comparison to the number of available gene sequences.