Title: Predicting multiple binding modes using a kernel method based on a Vector Space Model Molecular Descriptor

Authors: Forbes J. Burkowski, William W.L. Wong

Addresses: The David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada. ' The David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada

Abstract: We describe the use of our Vector Space Model Molecular Descriptor (VSMMD), based on a Vector Space Model (VSM) that is suitable for kernel studies in Quantitative Structure-Activity Relationship (QSAR) modelling. Our experiments provide convincing comparative empirical evidence that this kernel method can provide sufficient discrimination to predict various biological activities of a molecule with reasonable accuracy. Furthermore, together with a kernel feature space algorithm, experiments also provide convincing empirical evidence that our VSMMD can provide sufficient information to identify different binding modes with high accuracy.

Keywords: QSAR modelling; kernel methods; multiple binding modes; molecular descriptors; VSM; vector space models; quantitative structure-activity relationships; ligands.

DOI: 10.1504/IJCBDD.2009.027584

International Journal of Computational Biology and Drug Design, 2009 Vol.2 No.1, pp.58 - 80

Published online: 02 Aug 2009 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article