Among a wide range of alternative non-testing approaches for safety evaluation of chemicals, the <i>in silico</i> modeling new approach methodologies (NAMs) are viewed as a promising starting point for filling the existing gaps in safety and ecosafety data. As recently pointed out by the European Chemicals Agency and some authors, read-across is one of the most widely used alternative tools for hazard assessment, aimed at filling data gaps. Read-across not being a hardcore statistical approach, read-across-based predictions appear to be more appropriate for small data sets. Recently, we have reported a new similarity-based read-across algorithm for the prediction of toxicity (biological activity in general) of untested compounds from structural analogs (the tool available from https://sites.google.com/jadavpuruniversity.in/dtc-lab-software/home<u>)</u>. Three similarity estimation techniques such as Euclidean distance-based similarity, Gaussian kernel function similarity, and Laplacian kernel function similarity are used in this algorithm. The quality of predictions depends on the selection of the distance threshold, similarity threshold, and the number of most similar training compounds. The quality of quantitative predictions can be judged by using various correlation and error–based validation metrics. The tool is also able to handle graded data for classification-based predictions. The algorithm has now been used to predict various toxicity, ecotoxicity, and biological activity endpoints of query chemicals based on the chemical similarity approach. As the confidence of predictions for untested compounds is a piece of important information, we have addressed this issue by consideration of several criteria such as weighted standard deviation of the predicted values, coefficient of variation of the computed predictions, average similarity level of close training compounds for each query molecule, standard deviation and coefficient of variation of similarity levels, maximum similarity levels to positive and negative close training compounds, a perplexity measure indicating similarity to positive, negative or both classes of close training compounds, etc. The results will be discussed with several case studies of toxicity, ecotoxicity, and biological activity data sets. The derived reliability measures will provide greater confidence to the quality of predictions from the chemical read-across tool for new query compounds.

Chemical read-across: Quantitative predictions and their confidence measures

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.