Predicting In Vitro Neurotoxicity Induced by Nanoparticles Using Machine Learning

Abstract

The practice of non-testing approaches in nanoparticles hazard assessment is necessary to identify and classify potential risks in a cost effective and timely manner. Machine learning techniques have been applied in the field of nanotoxicology with encouraging results. A neurotoxicity classification model for diverse nanoparticles is presented in this study. A data set created from multiple literature sources consisting of nanoparticles physicochemical properties, exposure conditions and in vitro characteristics is compiled to predict cell viability. Pre-processing techniques were applied such as normalization methods and two supervised instance methods, a synthetic minority over-sampling technique to address biased predictions and production of subsamples via bootstrapping. The classification model was developed using random forest and goodness-of-fit with additional robustness and predictability metrics were used to evaluate the performance. Information gain analysis identified the exposure dose and duration, toxicological assay, cell type, and zeta potential as the five most important attributes to predict neurotoxicity in vitro. This is the first tissue-specific machine learning tool for neurotoxicity prediction caused by nanoparticles in in vitro systems. The model performs better than non-tissue specific models.

A Machine Learning Tool to Predict the Antibacterial Capacity of Nanoparticles

“…In addition, they can predict the impact of materials not yet synthesized, thereby contributing to safe-by-design approaches [ 51 ]. ML has been effectively employed for the prediction of toxicity profiles of NPs [ 52 , 53 , 54 , 55 , 56 ] and for the development of new antibiotics [ 57 , 58 ]. Furthermore, models for the prediction of the antimicrobial resistance for specific bacteria have been demonstrated [ 59 , 60 , 61 ].…”

NanoTox: Development of a parsimonious in silico model for toxicity assessment of metal-oxide nanoparticles using physicochemical features

“…A truly general model of nanoparticle cytotoxicity, independent of in vitro factors, would lead to significantly broader interpretability and wider applicability [43]. Our study departs also from the notion that tissue-specific models are superior to generalized models [38], and demonstrates that model interpretability is best achieved using a minimal non-redundant feature space, consistent with Occam’s parsimony. Furthre, with a view to increasing reliability, we have deployed the insights from our study into a majority-voting ensemble classifier.…”

“…Such techniques provide a non-invasive ‘instantaneous’ readout of nanoparticle toxicity [32–34], and originate from the evolution of QSAR models [35]. Machine learning models of nanoparticle toxicity have tended to be either generalized [36] or tissue-specific [37,38], and are built from experimental toxicity data that have been scored, standardised and curated into databases like the Safe and Sustainable Nanotechnology db (S2NANO) [39–41].…”

Prediction of the effective reproduction number of COVID-19 in Greece. A machine learning approach using Google mobility data

“…It also achieves the highest R 2 , Pearson’s and Spearman’s correlation values, outperforming in all metrics the other models. RF has been demonstrated to outperform other basic classifiers, even with missing values present in the dataset (Furxhi, Murphy et al 2019, Furxhi and Murphy 2020).…”