Prioritizing Direct Photolysis Products Predicted by the Chemical Transformation Simulator: Relative Reasoning and Absolute Ranking

• DOI: 10.1021/acs.est.0c08745
• Publication Date: 2021-04-23T01:23:39Z
• Authors: Chenyi Yuan; Caroline Tebes-St...; Eric J. Weber
• Keywords: 0301 basic medicine; Kinetics; 03 medical and health sciences; Photolysis; 13. Climate action; 01 natural sciences; Water Pollutants, Chemical; 6. Clean water; 0105 earth and related environmental sciences

The United States Environmental Protection Agency's Chemical Transformation Simulator (CTS) platform implemented the first freely available reaction library to predict direct photolysis products of organic contaminants in aquatic systems. However, initial version did not differentiate formation likelihood each predicted product, and therefore, number that are observed tended exponentially increase with prediction generation. To alleviate this problem, we employed relative reasoning algorithms remove unlikely products. We then ranked different schemes according their transformation kinetics removed slowly forming Applying two strategies improved precision (the percentage correctly over all products) by 34% 53% for internal evaluation set external set, respectively, when from three generations were considered. This also revealed new research directions improve predictions dominant phototransformation