A Quantitative Structure-Activity Relationship (QSAR) model was derived for
estimating the acute toxicity of pesticides against Lepomis macrochirus un
der varying experimental conditions. Chemicals were described by means of a
utocorrelation descriptors encoding lipophilicity (H-0 to H-5) and the H-bo
nding acceptor ability (HBA(0)) and H-bonding donor ability (HBD0) of the p
esticides. A three-layer feedforward neural network trained by the back-pro
pagation algorithm was used as statistical engine for deriving a powerful O
SAR model accounting for the weight of the fish, time of exposure, temperat
ure, pH, and water hardness.