If one has solved the structure of a known drug target, computational modelling and design approaches can be applied to ligand design to help guide the design of potential drug candidates which can then be tested in the laboratory. However, if one does not know the target structure the problem is more intractible. One expensive approach is to throw many compounds against the target in massive screening and combinatorial cehmistry experiments in the lab and then almost blindly sift out candidates with the most promising affinities. Apart from the significant cost of such efforts, one is literally searching for a needle in a haystack of trillions of billions of chemical possibilities. David Lloyd describes approaches developed by De Novo Pharmaceuticals (www.denovopharma.com) which can be used to tackle this situation in the computer laboratory whereby 2.5 million compounds can be screened per day on an 80-node computer farm.
The approach is to develop extended pharmacophore models which include both ligand and target site information based on a limited amount of experimental data. The best models may then be used in screening experiments against databases of compounds and when combined with de novo design techniques can be used to design new compounds which can then be tested in the laboratory. In his talk David Lloyd describes the development and validation of such approaches with a number of examples including the Estrogen receptor, HIV Protease, G-Protein Coupled Receptor (GPCR) targets, Histamine H3 receptor, and anti-cancer compounds against the poly-ADP-ribose polymerase (PARP) target.
You can listen to the full talk of David Lloyd with audio on the Cheminformatics & Modelling Community of Practice site at http://echeminfo.com/
Please note that David Lloyd has recently taken up the position of Hitachi Research Lecturer in the Department of Biochemistry, Trinity College Dublin, building up the Molecular Design Group in the department (http://www.tcd.ie/Biochemistry) and is coordinating the activities of the biomolecular modelling group in Trinity’s recently funded IITAC research programme, http://www.tchpc.tcd.ie/~iitac/index.php
Barry Hardy
www.douglasconnect.com
Comments