Clustering and classification techniques to assess aquatic toxicity

The goal of toxicity, prediction is to describe the relationship between chemical properties, on the one hand, and biological and toxicological processes, on the other. Knowledge about the causes of toxicity is incomplete. No single property can satisfy the requirement to model the toxic activity. In the present study we consider different methods to build up models useful for aquatic toxicity prediction. Our study is in the tradition of SAR and QSAR methods, but tries to predict a category. Due to the variability of the toxicity phenomenon, classification methods may present advantages because they refer to intervals of the observed toxic effect. Furthermore classification of compounds according to their toxicity has direct application for regulation of chemicals. In the paper we will report results obtained from the preparation and study of a data set of different classes of chemicals; starting from recursive partitioning algorithms we will test their results against clustering and classifiers.