Exploiting structural genomics information to corporate protein flexibility in drug design

General info

Date from - to: 01 Jan 2005 - 01 Oct 2009
Project leader(s): Vlieg, de Jacob Prof. dr.

Abstract

Structural bioinformatics allows, in principle, the design of small molecules, such as potential pharmaceutical drugs, that bind specifically and tightly to a protein target. However, a protein may change its shape in such an interaction, and such changes are extremely difficult to predict. We need to know more about protein flexibility in order to understand how drugs exert their biological effects. In this project we focus on predicting the conformational changes, or alternatively predicting the selection of protein conformations, that occur in a protein upon binding with a ligand, in particular at the binding site. The compilation of structures of proteins within the same family is thought to be instrumental in understanding protein flexibility.
Software has been developed and installed to simulate and test protocols for ligand inducible binding sites using molecular dynamics simulations with hydrophobic probes. A protocol has been developed and tested to incorporate induced-fit in protein-ligand docking with the help of the FlexE and Yasara programs. The DFIRE potential, a scoring function that allows us to filter conformations, has been incorporated into existing software (Rotacal - capable of generating different versions of the active site based on side chain flexibility (rotamers)- and molden) and evaluated. The results for the Nuclear Receptor Molecular Database Systems For Automatic Collection And Interpretation Of Chemical And Biological Information have been reported, both as a publication and as a conference presentation.