183 Prediction of acute aquatic toxicity to fish comparing different QSAR approaches

The proposal of architecture for chemical splitting to optimize QSAR models for aquatic toxicity

One of the challenges in the field of quantitative structure-activity relationship (QSAR) analysis is the correct classification of a chemical compound to an appropriate model for the prediction of activity. Thus, in previous studies, compounds have been divided into distinct groups according to their mode of action or chemical class. In the current study, theoretical molecular descriptors were used to divide 568 organic substances into subsets with toxicity measured for the 96-h lethal median concentration for the Fathead minnow (Pimephales promelas). Simple constitutional descriptors such as the number of aliphatic and aromatic rings and a quantum chemical descriptor, maximum bond order of a carbon atom divide compounds into nine subsets. For each subset of compounds the automatic forward selection of descriptors was applied to construct QSAR models. Significant correlations were achieved for each subset of chemicals and all models were validated with the leave-one-out internal validation procedure (R(2)(cv) approximately 0.80). The results encourage to consider this alternative way for the prediction of toxicity using QSAR subset models without direct reference to the mechanism of toxic action or the traditional chemical classification.

Predictive Models for Aquatic Toxicity of Aldehydes Designed for Various Model Chemistries

Predictive models for the aquatic toxicity of aldehydes were designed for a set of 50 aromatic or aliphatic compounds containing at least one aldehyde group, for which the acute toxicity data for the fathead minnow (Pimephales promelas) are available (96 h test assessing 50% lethal waterborne concentration). The molecular descriptors were based on calculations with various semiempirical or ab initio model chemistries. The resulting four-parameter models were evaluated according to the correlation coefficients R(2). The best predictive model was obtained with the HF/STO-3G model chemistry (R(2) = 0.868), while the models designed for descriptors based on ab initio calculations of higher level showed a slightly worse predictivity (the HF/3-21G(d) based model R(2) = 0.800, the HF/6-31G(d) based model R(2) = 0.808, the B3LYP/6-31G(d,p) based model R(2) = 0.812). With the semiempirical methods a good predictivity was observed with the PM3 based model (R(2) = 0.811) and the AM1 based model (R(2) = 0.791), but the MNDO based model showed the worst predictivity (R(2) = 0.760). In all ab initio models and the PM3 model very similar descriptors were involved. The importance of the descriptor logarithm of the partition coefficient logP for toxicity prediction was confirmed. Additionally, the descriptors encoding the negatively charged molecular surface area, hydrogen bonding molecular surface area, and reactivity of aldehyde group were identified as essential for the toxicity prediction of aldehydes.