In recent years, the availability of high resolution structures of enzyme-inhibitor complexes have led to an increased understanding of molecular interactions. Using this structural information, computer-based approaches help identify or design ligands that posses good steric and chemical complementarity to various sites on the enzyme. This process is referred to as """"""""structure-based molecular design"""""""". The long term objective of this work is to develop a method tocomputationally screen and evaluate ligands as potential lead compounds. We are seeking a method that is rapid and reasonably accurate. The DOCK suite of programs, developed by the Kuntz group, identifies and characterizes sites on a receptor or protein for inhibitor binding; orients a database of molecules; and evaluates these molecules for goodness of fit. Site identification and characterization has traditionally be done by the program SPHGEN. More recently, we have developed a new method, called SURFSPH, which identifies putative atom positions on the protein. This program can also been used to help locate differences between enzymes from different species and as an aid in ligand modification. MidasPlus has been used here to visualize and display these spheres (see 1). Additionally, MidasPlus is used--along with the midas delegate, Viewdock, to examine different conformations and orientations of ligands from the DOCK output. This is a continuation of the work which identified inhibitors of the enzyme, xanthine-hypoxanthine-guanine-phosphoribosyl transferase, xhg-ptrase.
Showing the most recent 10 out of 508 publications
