Applications of QSARs in toxicology: a US Government perspective

Application of a genetic algorithm and an artificial neural network for global prediction of the toxicity of phenols to Tetrahymena pyriformis

ABSTRACT

Genetic algorithm (multiparameter linear regression; GA-MLR) and genetic algorithm-artificial neural network (GA-ANN) global models have been used for prediction of the toxicity of phenols to Tetrahymena pyriformis. The data set was divided into 150 molecules for training, 50 molecules for validation, and 50 molecules for prediction sets. A large number of descriptors were calculated and the genetic algorithm was used to select variables that resulted in the best-fit to models. The six molecular descriptors selected were used as inputs for the models. The MLR model was validated using leave-one-out, leave-group-out cross-validation and external test set. A three-layered feed forward ANN with back-propagation of error was generated using six molecular descriptors appearing in the MLR model. Comparison of the results obtained using the ANN model with those from the MLR revealed the superiority of the ANN model over the MLR. The root mean square error of the training, validation, and prediction sets for the ANN model were calculated to be 0.224, 0.202, and 0.224 and correlation coefficients (r2) of 0.926, 0.943, and 0.925 were obtained. The improvements are because of non-linear correlations of the toxicity of the compounds with the descriptors selected. The prediction ability of the GA-ANN global model is much better than that of previously proposed models.

GRAPHICAL ABSTRACT

Application of QSARs and VFARs to the rapid risk assessment process at US EPA

With continued development of new chemicals and genetically engineered microbes as potential agents for terrorism and industrial development, there is a great need for the continued development and application of quantitative structure activity relationships (QSARs) and virulence factor activity relationships (VFARs). Development and application of QSARs and VFARs will facilitate efficient and streamlined use of dwindling resources and assessment of risks associated with exposures to chemical and biological agents. To facilitate the continued development of QSARs and VFARs at US Environmental Protection Agency, a two day workshop was organized June 20-21, 2006, in Cincinnati, OH, USA. This article summarizes the workshop report by highlighting the importance of continued QSAR research, the current state of VFAR science, and the guidance provided to the National Homeland Security Research Center and National Risk Management Research Laboratory by an expert panel for the continued use and development of computational approaches.

A new approach to QSAR modelling of acute toxicity

A new QSAR approach based on a Quantitative Neighbourhoods of Atoms description of molecular structures and self-consistent regression was developed. Its prediction accuracy, advantages and limitations were analysed from three sets of published experimental data on acute toxicity: 56 phenylsulfonyl carboxylates for Vibrio fischeri; 65 aromatic compounds for the alga Chlorella vulgaris and 200 phenols for the ciliated protozoan Tetrahymena pyriformis. According to our findings, the proposed approach provides a good correlation and prediction accuracy (r(2) = 0.908 and Q(2) = 0.866) for the set of 56 phenylsulfonyl carboxylates and the 65 aromatic compounds tested on C. vulgaris (r(2) = 0.885, Q(2) = 0.849). For the 200 phenols tested on T. pyriformis, the prediction accuracy was r(2) = 0.685 and Q(2) = 0.651. This is at least as good as the best results obtained with the other QSAR methods originally used on the same data sets.

Computer-based QSARs for predicting mixture toxicity of benzene and its derivatives

During the past decades, the Quantitative structure-activity relationships (QSARs) have been proven to be reliable tools when little or no empirical data are available in medicinal chemistry, biochemistry, toxicology, and environmental sciences. However, only few studies that quantitatively predict mixtures toxicity have been reported. In this study, the QASR models for the binary mixtures toxicity of 12 benzene and its derivatives, including eight non-polar-narcotic compounds and four polar narcotic compounds were developed, without reference to exact toxicity mechanisms of single compounds. All parameters for the QSAR studies were defined on the basis of quantum mechanical calculations and these parameters were selected by the stepwise procedure. The results of this study provided a simple means of predicting the binary mixtures toxicity from the chemical structure.

Statistically validated QSARs, based on theoretical descriptors, for modeling aquatic toxicity of organic chemicals in Pimephales promelas (fathead minnow)

The use of Quantitative Structure-Activity Relationships in assessing the potential negative effects of chemicals plays an important role in ecotoxicology. (LC50)(96h) in Pimephales promelas (Duluth database) is widely modeled as an aquatic toxicity end-point. The object of this study was to compare different molecular descriptors in the development of new statistically validated QSAR models to predict the aquatic toxicity of chemicals classified according to their MOA and in a unique general model. The applied multiple linear regression approach (ordinary least squares) is based on theoretical molecular descriptor variety (1D, 2D, and 3D, from DRAGON package, and some calculated logP). The best combination of modeling descriptors was selected by the Genetic Algorithm-Variable Subset Selection procedure. The robustness and the predictive performance of the proposed models was verified using both internal (cross-validation by LOO, bootstrap, Y-scrambling) and external statistical validations (by splitting the original data set into training and validation sets by Kohonen-artificial neural networks (K-ANN)). The model applicability domain (AD) was checked by the leverage approach to verify prediction reliability.

Ranking of aquatic toxicity of esters modelled by QSAR

Alternative methods like predictions based on Quantitative Structure-Activity Relationships (QSARs) are now accepted to fill data gaps and define priority lists for more expensive and time consuming assessments. A heterogeneous data set of 74 esters was studied for their aquatic toxicity, and available experimental toxicity data on algae, Daphnia and fish were used to develop statistically validated QSAR models, obtained using multiple linear regression (MLR) by the OLS (Ordinary Least Squares) method and GA-VSS (Variable Subset Selection by Genetic Algorithms) to predict missing values. An ESter Aquatic Toxicity INdex (ESATIN) was then obtained by combining, by PCA, experimental and predicted toxicity data, from which model outliers and esters highly influential due to their structure had been eliminated. Finally this integrated aquatic toxicity index, defined by the PC1 score, was modelled using only a few theoretical molecular descriptors. This last QSAR model, statistically validated for its predictive power, could be proposed as a preliminary evaluative method for screening/prioritising esters according to their integrated aquatic toxicity, just starting from their molecular structure.

POPs: a QSAR system for developing categories for persistent, bioaccumulative and toxic chemicals and their metabolites

This paper presents the framework of a QSAR-based decision support system which provides a rapid screening of potential hazards, classification of chemicals with respect to risk management thresholds, and estimation of missing data for the early stages of risk assessment. At the simplest level, the framework is designed to rank hundreds of chemicals according to their profile of persistence, bioaccumulation potential and toxicity often called the persistent organic pollutant (POP) profile or the PBT (persistent bioaccumulative toxicant) profile. The only input data are the chemical structure. The POPs framework enables decision makers to introduce the risk management thresholds used in the classification of chemicals under various authorities. Finally, the POPs framework advances hazard identification by integrating a metabolic simulator that generates metabolic map for each parent chemical. Both the parent chemicals and plausible metabolites are systematically evaluated for metabolic activation and POPs profile.

A protocol to select high quality datasets of ecotoxicity values for pesticides

The key to any QSAR model is the underlying dataset. In order to construct a reliable dataset to develop a QSAR model for pesticide toxicity, we have derived a protocol to critically evaluate the quality of the underlying data. In developing an appropriate protocol that would enable data to be selected in constructing a QSAR, we concentrated on one toxicity end point, the 96 h LC50 from the acute rainbow trout study. This end point is key in pesticide regulation carried out under 91/414/EEC. The dataset used for this exercise was from the US EPA-OPP database.

Assessment and Modeling of the Toxicity of Organic Chemicals to Chlorella vulgaris: Development of a Novel Database

This study reports a database of toxicity values for 91 compounds assessed in a novel, rapid, and economical 15 min algal toxicity test. The toxicity data were measured using the unicellular green alga Chlorella vulgaris in an assay that determined the disappearance of fluorescein diacetate. The chemicals tested covered a wide range of physicochemical properties and mechanisms of action. Quantitative activity-activity relationships with the toxicity of the chemicals to other species (Tetrahymena pyriformis, Vibrio fischeri, and Pimephales promelas) showed strong relationships, although some differences resulting from different protocols were established. Quantitative structure-activity relationships (QSARs) were determined using linear [multiple linear regression (MLR)] and nonlinear [k-nearest neighbors (KNN)] methods. Three descriptors, accounting for hydrophobicity, electrophilicity, and a function of molecular size corrected for the presence of heteroatoms, were found to be important to model toxicity. The predictivity of MLR was compared to KNN using leave-one-out cross-validation and the simulation of an external test set. MLR demonstrated greater stability in validation. The results of this study showed that method selection in QSAR is task-dependent and it is inappropriate to resort to more complicated but less transparent methods, unless there are clear indications (e.g., inability of MLR to deal with the data set) for the need of such methods.

QSAR Analysis of the Toxicity of Aromatic Compounds to Chlorella vulgaris in a Novel Short-Term Assay

The use of alternative toxicity tests and computational prediction models is widely accepted to fill experimental data gaps and to prioritize chemicals for more expensive and time-consuming assessment. A novel short-term toxicity test using the alga Chlorella vulgaris was utilized in this study to produce acute aquatic toxicity data for 65 aromatic compounds. The compounds tested included phenols, anilines, nitrobenzenes, benzaldehydes and other poly-substituted benzenes. The toxicity data were employed in the development of quantitative structure-activity relationships (QSARs). Using multiple regression (MLR) and partial least squares (PLS) analyses, statistically significant, transparent and interpretable QSARs were developed using a small number of physicochemical descriptors. A two-descriptor model was developed using MLR (log(1/EC50)=0.73 log Kow-0.59 Elumo-1.91; n=65, r2=0.84, r2CV=0.82, s=0.43) and a four-descriptor model using PLS (log(1/EC50)=0.40 log Kow-0.23 Elumo+9.84 Amax+0.20 0chiv-5.40; n=65, r2=0.86, q2=0.84, RMSEE=0.40). The latter model was obtained by stepwise elimination of variables from a set of 102 calculated descriptors. Both models were validated successfully by simulating external prediction through the use of complementary subsets. The two factors, which were identified as being critical for the acute algal toxicity of this set of compounds were hydrophobicity and electrophilicity.