A general QSAR model for predicting the toxicity of organic chemicals to luminescent bacteria (Microtox test)
Industrial countries today are facing serious ecotoxicological problems resulting from the release of xenobiotics to the environment. The Microtox (registered trademark) test is now widely used to estimate the toxicity of chemicals in screening procedures and derive quantitative structure activity relationship (QSAR) models by means of regression analysis. In the study reported here a large data set of Microtox toxicity results was used to derive a general QSAR model. Chemicals were described by means of a modified autocorrelation method. The autocorrelation vectors were generated from atomic contributions encoding the hydrophobicity and molar refractivity of the molecules. A three layer backpropagation neural network was used to design the model. The obtained results were compared with those obtained from a principal components regression analysis.
Bibliographic Reference: Article: SAR and QSAR in Environmental Research, Vol. 4 (1995) No. 1, pp. 29-38
Availability: Available from Dr J Devillers, CTIS, 21 rue de la Bannière, 69003 Lyon (FR)
Record Number: 199610306 / Last updated on: 1996-03-29
Original language: en
Available languages: en