S. Anzali, J. Gasteiger, U. Holzgrabe, J. Polanski, J. Sadowski, A. Teckentrup, M. Wagener
"The Use of Self-Organizing Neural Networks in Drug Design"
in "3D QSAR im Drug Design - Volume 2"
H. Kubinyi, G. Folkers, Y. C. Martin (Ed.), Kluwer/ESCOM, Dordrecht, NL, 1998, p. 273-299

The Use of Self-Organizing Neural Networks in Drug Design

1. Introduction

2. Self-Organizing Neural Networks

3. Computational Details

4. One Molecule into One Network

4.1. Maps of molecular surfaces
4.1.1. Methods and results
4.1.2. Visual comparison of Kohonen maps
4.1.3. Averaged maps
4.1.4. Maps as a two-dimensional representation of molecules
4.1.5. Bioisosteric design
4.2. Comparative maps
4.2.1. The template approach
4.2.2. Shape analysis of CBG ligands
4.2.3. Backprojection of maps onto molecular shapes
4.2.4. Descriptors from comparative maps
4.3. Quantitative structure-activity studies

5. Several Molecules into One Network

5.1. Representation of molecules
5.2. Topological autocorrelation and the location of biologically active compounds
5.3. Autocorrelation of molecular surfaces properties
5.4. Modelling CBG affinity by a combination of two different neural networks
5.5. Modeling of chemical libraries
5.5.1. Comparison of the xanthene, the cubane and the adamantane libraries
5.5.2. Deconvolution of xanthene sublibraries

6. Conclusion and Outlook