Support Vector Machine (SVM) as Alternative Tool to Assign Acute Aquatic Toxicity Warning Labels to Chemicals

Abstract

Quantitative structure-activity relationship (QSAR) analysis has been frequently utilized as a computational tool for the prediction of several eco-toxicological parameters including the acute aquatic toxicity. In the present study, we describe a novel integrated strategy to describe the acute aquatic toxicity through the combination of both toxicokinetic and toxicodynamic behaviors of chemicals. In particular, a robust classification model (TOXclass) has been derived by combining Support Vector Machine (SVM) analysis with three classes of toxicokinetic-like molecular descriptors: the autocorrelation molecular electrostatic potential (autoMEP) vectors, Sterimol topological descriptors and logP(o/w) property values. TOXclass model is able to assign chemicals to different levels of acute aquatic toxicity, providing an appropriate answer to the new regulatory requirements. Moreover, we have extended the above mentioned toxicokinetic-like descriptor set with a more toxicodynamic-like descriptors, as for example HOMO and LUMO energies, to generate a valuable SVM classifier (MOAclass) for the prediction of the mode of action (MOA) of toxic chemicals. As preliminary validation of our approach, the toxicokinetic (TOXclass) and the toxicodynamic (MOAclass) models have been applied in series to inspect both aquatic toxicity hazard and mode of action of 296 chemical substances with unknown or uncertain toxicodynamic information to assess the potential ecological risk and the toxic mechanism.

Predicting the Thermal Stability of Nitroaromatic Compounds Using Chemoinformatic Tools

“…[5,6] These principles have already been applied in the last years for few QSPR models to the evaluation of substances toxicity in order to circumvent the ethical problems related to the animal testing. [7,8] While the popularity of QSAR methods can be certainly acknowledged within the field of biology [9] , pharmaceutics [10,11] , and toxicity predictions [12][13][14][15] , the QSPR approach may find significant applications within the REACH framework for charAbstract: In the framework of the European REACH regulation major attention was recently devoted to toxicological and ecotoxicological problems while little attention has been dedicated to other important applications concerning chemical hazards, for instance, explosive properties. In this work different chemoinformatic tools such as partial least squares, multilinear regressions, and decision trees have been used for the development of a novel quantitative structure-property relationships to predict the heat of decomposition of a series of nitroaromatic compounds.…”

Multi‐class Mode of Action Classification of Toxic Compounds Using Logic Based Kernel Methods

“…We now compare the performance of DL_SVILP with a recent approach proposed by Michielan et al [23] The study applied multi-class classification method, namely one-vsone SVMs by using the software LIBSVM. [35] The compounds were represented by toxicokinetic-like and toxicodynamiclike descriptors.…”

“…The experimental results demonstrate the efficacy of our approach. The technique outperforms the other approaches in the study and the method presented by Michielan et al [23] 2 Multi-class Computational Methods…”

“…Number of compounds and accuracy for the one-vs-one SVMs [14] Accuracy values are calculated by analyzing the data presented with the paper. [23] Class …”

“…[19][20][21] In contrast to machine learning methods expert systems have mostly been utilized to predict toxicity of compounds. [22] In a recent study Michielan et al [23] presented an approach to classify the mode of action of 292 compounds in EPAFHM database. The study applied multi-class SVMs in conjunction with molecular descriptors including autocorrelation molecular electrostatic potential vectors, [24] Sterimol topological values, [25] logP numbers and quantum-physicalchemical properties (hydrogen bonding donors and acceptors, topological polar and approximate surface areas, HOMO, the difference between HOMO and LUMO values).…”