Evaluation of High-Throughput Chemical Functional Use Models
Katherine Phillips, Kristin Isaacs, Daniel T. Chang, Kristan J. Markey · ISEE Conference Abstracts · 2018
The United States Environmental Protection Agency’s Office of Research and Development has generated high-throughput quantitative structure-use relationship (QSUR) models that predict the probability of a chemical fulfilling a specific functional use in consumer products and chemical processes, through knowledge of that chemical’s structure. Identification of a chemical’s functional use provides clues to both its expected concentration and the intentionality of its presence in a product. Information on the presence or amount of chemicals in products can be used as inputs to exposure models but is often lacking. For these data-poor chemicals, predictions of functional use can be used to parameterize high-throughput exposure models and support rapid prioritization of chemicals for risk and alternatives assessments. The QSUR models were built using a training set of functional use data derived primarily from the consumer product sector. Publicly available functional uses reported for chemicals in the industrial sector by manufacturers under the Chemical Data Reporting (CDR) Rule of the Toxic Substances Control Act were used to evaluate the QSUR model predictions for industrial chemicals. A total of 5153 chemicals in CDR had reported functional uses that could be compared to QSUR predictions. Only 2689 of these are single compounds with a definite, disclosed structure. Of those 2689, 1603 had predictions with a high probability (≥80%) of having a functional use and were within a model’s domain of applicability. In addition, sector information (industrial or consumer) reported in the CDR was used to explore structural differences in chemicals having the same function in different sectors showing that chemicals with certain functional uses (solvents, preservatives) can differ drastically in structure from one sector to another, while others (emulsifiers, surfactants) can have similar structures in both sectors. This abstract does not necessarily reflect US EPA policy.