Quantum mechanics descriptors in a nano-QSAR model to predict metal oxide nanoparticles toxicity in human keratinous cells

The nanosafety assessment of ENMs under a dermal exposure scenario: from key molecular events toin silicomodeling tools

“…7), nano-(Q)SAR enables the given toxic effects of ENMs to be determined by their 197 With the EU ban on animal testing, (Q)SAR has been extensively used as an alternative approach in mechanistic interpretation, tiered testing, grouping, and ranking the toxic potency of ENMs for risk assessment. Table 5 summarizes recent (Q)SAR studies regarding the cytotoxicity (cellular uptake and HaCaT cell viability) and/or genotoxicity (results of the bacterial reverse mutation test) of MeOx NPs, 112,[198][199][200][201][202][203][204][205][206][207][208][209] carbon nanotubes 210,211 and fullerenes. 212,213 Datasets for (Q)SAR modeling can be obtained from literature, databases or experiments and should contain sufficient chemically diverse data.…”

“…197 With the EU ban on animal testing, (Q)SAR has been extensively used as an alternative approach in mechanistic interpretation, tiered testing, grouping, and ranking the toxic potency of ENMs for risk assessment. Table 5 summarizes recent (Q)SAR studies regarding the cytotoxicity (cellular uptake and HaCaT cell viability) and/or genotoxicity (results of the bacterial reverse mutation test) of MeOx NPs, 112,198–209 carbon nanotubes 210,211 and fullerenes. 212,213…”

Converting Nanotoxicity Data to Information Using Artificial Intelligence and Simulation

“…For large (i.e., size >5 nm) and polydisperse materials like NMs, quantum chemical calculations require large computing resources. In some studies, NMs were represented by unit cells or small-sized clusters to make the quantum chemical calculations tractable. , However, quantum chemical nanodescriptors are only useful for idealized, pristine monodisperse NMs and are not applicable to account for distributions of sizes, shapes, and surface modifications. On the other hand, although quantum chemical descriptors can provide electronic properties of a molecule, the need for high computational resources limits their application in machine learning .…”

“…There are many open-source and commercial software packages, such as VASP (Vienna Ab initio Simulation Package) and Gaussian, that contain semiempirical or ab initio methods. , They can provide information about geometric and electronic properties, such as energies of the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). These have been widely used as features for nanotoxicity modeling. ,− For example, an MLR model was established to describe the toxicity of 17 MONPs to Escherichia coli using the calculated quantum chemical descriptors. The Δ H Me+ (i.e., the enthalpy of formation of a gaseous cation having the same oxidation state as that in the metal oxide structure) was found to be useful to predict bacterial toxicity .…”

“…Models that clearly explain how they produce predictions are also called glass-box or white-box ML models. The simplest ML algorithm to develop intrinsic nanotoxicity models is MLR. ,,,,− This method uses property and structural features as regressors to construct a linear relationship equation that predicts nanotoxicity . Such models are easily interpretable because the signs and magnitudes of the regression coefficients show how each feature influences nanotoxicity.…”