Molecular similarity analysis as tool to prioritize research among emerging contaminants in the environment

The very large number of emerging contaminants entering the water supply makes it desirable to assess these chemicals’ environmental fate and behavior without direct measurements. Structure–property prediction models are promising in this regard, and quantitative molecular similarity assessment (QMSA) is one particularly appealing option. This study has two objectives related to QMSA modeling of emerging contaminants: (1) demonstrating that QMSA models can be used to accurately predict an environmental engineering parameter of interest, e.g., in vitro estrogenicity measurements, for highly diverse chemical classes; and (2) assessing the extent to which QMSA approaches can be used to prioritize among unmeasured chemicals and determine which additional measurements will result in maximally increased model accuracy. The results of this study are promising in both regards. QMSA models were found to predict the test parameter, estrogenicity, with cross validation coefficients (q2) as high as 0.84. Results of a paired t-test to evaluate the difference in increased model accuracy associated with QMSA-selection versus random-selection of additional compounds yielded statistically significant P-values < 0.0001. Taken together, the results of this study suggest that QMSA could dramatically reduce the number of laboratory and field measurements required to characterize as-yet unknown environmental fate and behavior parameters for diverse emerging contaminants of regulatory interest. Additional preliminary work points to the promise of QMSA for prediction of fundamental adsorption parameters; e.g., distribution coefficient (Kd) for selected pharmaceuticals onto activated sludge during municipal wastewater treatment.

Figure optionsDownload as PowerPoint slideHighlights
► A quantum molecular similarity assessment (QMSA) model makes accurate predictions for diverse chemicals.
► QMSA prioritization of unmeasured chemicals identifies representative compounds in a dataset.
► QMSA predicts distribution coefficient (Kd) for pharmaceuticals onto wastewater sludge.