Engineered nanomaterials (ENMs) are ubiquitous nowadays, finding their application in different fields of technology and various consumer products. Virtually any chemical can be manipulated at the... Show moreEngineered nanomaterials (ENMs) are ubiquitous nowadays, finding their application in different fields of technology and various consumer products. Virtually any chemical can be manipulated at the nano-scale to display unique characteristics which makes them appealing over larger sized materials. As the production and development of ENMs have increased considerably over time, so too have concerns regarding their adverse effects and environmental impacts. It is unfeasible to assess the risks associated with every single ENM through in vivo or in vitro experiments. As an alternative, in silico methods can be employed to evaluate ENMs. To perform such an evaluation, we collected data from databases and literature to create classification models based on machine learning algorithms in accordance with the principles laid out by the OECD for the creation of QSARs. The aim was to investigate the performance of various machine learning algorithms towards predicting a well-defined in vivo toxicity endpoint (Daphnia magna immobilization) and also to identify which features are important drivers of D. magna in vivo nanotoxicity. Results indicated highly comparable model performance between all algorithms and predictive performance exceeding ∼0.7 for all evaluated metrics (e.g. accuracy, sensitivity, specificity, balanced accuracy, Matthews correlation coefficient, area under the receiver operator characteristic curve). The random forest, artificial neural network, and k-nearest neighbor models displayed the best performance but this was only marginally better compared to the other models. Furthermore, the variable importance analysis indicated that molecular descriptors and physicochemical properties were generally important within most models, while features related to the exposure conditions produced slightly conflicting results. Lastly, results also indicate that reliable and robust machine learning models can be generated for in vivo endpoints with smaller datasets. Show less
Assessing the risks of engineered nanomaterials (ENMs) solely on the basis of experimental assays is time-consuming, resource intensive, and constrained by ethical considerations (such as the... Show moreAssessing the risks of engineered nanomaterials (ENMs) solely on the basis of experimental assays is time-consuming, resource intensive, and constrained by ethical considerations (such as the principles of the 3Rs of animal testing). The adoption of computational toxicology in this field is a high priority. Computational toxicology is able to contribute to the prediction of the extent of toxic effects of untested ENMs, to the hazard categorization and labeling of ENMs, and to the establishment of hazard threshold values that are sufficiently protecting the ecosystem with respect to the ENMs of concern. These three steps are listed by the European Chemicals Agency (ECHA) as the three elements in evaluating the hazards of ENMs. This study has expanded the use of computational toxicology in the hazard assessment with regard to the safe handling of ENMs. The results obtained contribute to the integration and evaluation of toxicity data, the identification of research gaps on ENM-related modeling, and the development of nano-SARs and SSDs for metallic ENMs. Despite the uncertainties that are associated with our results, as mainly due to limited data quality and availability, we managed to take this field one step forwards and contribute to better-informed regulatory decisions of ENMs. Show less