Research Output per year
The Kohonen self-organizing maps technique is used to project non-linearly tridimensional molecular properties such as the electrostatic potential and the hydrophobicity of various ligands into a bidimensional space conserving the topology of the information. The aim of this work is to compare simultaneously the ligands in order to determine the required elements for their high activity and selectivity.
|Effective start/end date||1/11/97 → 31/12/02|
- thromboxane A2 receptors
- molecular electrostatic potential
- 5-HT1D receptors
- Kohonen mapping
- artificial neural networks
Application of a Kohonen Neural Network to the Analysis of Data Regarding the Alkylation of Toluene with Methanol Catalyzed by ZSM-5 Type ZeolitesPetit, J., Zupan, J., Leherte, L. & Vercauteren, D. P., 1 Nov 2002, In : Computers and Chemistry. 26, 6, p. 557-572 16 p.
Research output: Contribution to journal › Article