Quasi-SMILES as a tool to utilize eclectic data for predicting the behavior of nanomaterials

Abstract

Nowadays, nanomaterials are often considered a scientific hit. However, despite the immense advantages of nanomaterials, there are studies, which have shown that these materials can also harmfully impact both human health and the environment. A preliminary evaluation of the hazards related to nanomaterials can be performed using predictive models. The aim of the present study is building up a single QSAR model for predicting cytotoxicity of metal oxide nanoparticles on (i) Escherichia coli (E. coli) and (ii) human keratinocyte cell line (HaCaT) based on the representation of the available eclectic data, encoded into quasi-SMILES. Quasi-SMILES is an analogue and an attractive alternative of traditional simplified molecular input-line entry systems (SMILES). In contrast to traditional SMILES quasi-SMILES are a tool to represent not only molecular structures, but also different conditions, such as physicochemical properties and experimental conditions. The statistical quality of the models are average correlation coefficient (r 2) and root mean squared error (RMSE) for the training set 0.79 and 0.216; the average r 2 and RMSE for validation set are 0.90 and 0.247, respectively.

Towards the Development of Global Nano-Quantitative Structure–Property Relationship Models: Zeta Potentials of Metal Oxide Nanoparticles

“…It should be noted that the quality of any predictive model is the ability to adequately predict endpoints for external objects. In that case, the external prediction is invisible during model development [ 37 ]. At the same time, an excellent statistical quality of a model for the training set is often an indicator of overfitting [ 38 ].…”

“…In the presented case, the optimal descriptor is a translator of eclectic information into the predictive model [ 37 , 38 , 39 ]. For instance, using Scheme 1 , Al 2 O 3 nanoparticles form the first row in Table 1 (a nominal size of 11.40 nm and a size in media of 94.70 nm) were attributed to the quasi-SMILE code O=[Al]O[Al]=O%11%51.…”

QSPR/QSAR: State-of-Art, Weirdness, the Future

“…It is to be noted, however, in some cases, the molecular structure is not informative to build up a predictive model of endpoints [81][82][83][84][85][86][87][88][89][90][91][92][93][94][95]. Meanwhile, the definition of a model as a mathematical function of experimental conditions (after consultations with experimentalists) is a shorter and consequently more attractive way to solve the corresponding task.…”

Nano-(Q)SAR for Cytotoxicity Prediction of Engineered Nanomaterials

“…A single QSAR model for predicting cytotoxicity of 16 metal oxide NPs both towards E. coli and HaCaT cells was built in [37]. The model was based on the representation of the available data, encoded into quasi-SMILES.…”

“…Quasi-SMILES are a tool to represent different conditions: Physico-chemical properties and experimental conditions. The statistical quality of the models was evaluated using average determination coefficient R 2 and root mean squared error (RMSE) for the training set, which were equal to 0.79 and 0.216; R 2 and RMSE for the validation set were equal to 0.90 and 0.247, respectively [37].…”