2nd Year PhD Student
|
Alasdair Laurie
2nd Year PhD Student
|
|
|
Bioinformatics Group
|
Phone
Fax
E-mail
|
Having already created Q-SiteFinder to detect binding sites on proteins, the task now remains to develop web-based virtual screening and de novo design tools that use the predicted binding sites. An automated de novo drug design tool is currently in development.
The six month research project associated with this course involved the development
of Q-SiteFinder. Q-SiteFinder is an energy-based method of ligand binding site prediction. It works
by creating a three dimensional grid around a protein and placing a methly probe at each grid point.
The interaction energy of each probe site with the protein is estimated using the GRID forcefield.
Probes with a favourable interaction energy are retained and clustered according to their
spatial proximity. Clusters are ranked according to their total interaction energies.
The lowest energy cluster is the first predicted binding site. Q-SiteFinder was found to predict
sites with a higher precision than Pocket-Finder, a pocket detection algorithm.
Laurie ATR & Jackson RM (2005). Q-SiteFinder: an energy-based method for the prediction of protein-ligand binding sites. Bioinformatics, 21:1908-1916.
Leslie, M (ed.) (2005). Death by Design. Science, Vol. 305, No. 5275, Page 1093. Reviews my apoptosis website.
Laurie ATR & Jackson RM (2005). Online Tools for Structure Based Drug Design. Young Modellers Forum, London, 02/12/2005. Oral Presentation.
Laurie ATR & Jackson RM (2005). Q-SiteFinder: An online tool for ligand-protein binding site prediction. Proceedings of the Leeds Annual Statistical Research Workshop, 24:150-151. Poster Presentation.