Prediction Data Available: Predictive Models for Acute Oral Toxicity
The prediction data set is now available for the ICCVAM-sponsored global project to develop in silico models of acute oral systemic toxicity. The goal of this project is to develop in silico models that will predict five specific endpoints needed by regulatory agencies. Endpoints of interest include identification of “very toxic” chemicals (LD50 less than 50 mg/kg), “nontoxic” chemicals (LD50 greater than or equal to 2000 mg/kg), point estimates for rodent LD50s, and categorization of toxicity hazard using the U.S. Environmental Protection Agency (EPA) and Globally Harmonized System of Classification and Labelling (GHS) classification schemes. Modelers may build predictive models for any or all of these endpoints.
Models will be developed and evaluated using rat acute oral toxicity data collected by NICEATM and the EPA National Center for Computational Toxicology. Models meeting criteria defined by the project organizing committee will be used to generate consensus predictions for the acute oral toxicity endpoints of interest. A summary of the project and developed models will be submitted for publication in the peer-reviewed literature, and the toxicity predictions generated by the models will be made available via EPA’s Chemistry Dashboard.
Resources available on the project page include data files, timeline, and a downloadable document that specifies project objectives and scope, details on the data and processing steps, model evaluation criteria, and additional considerations for project participants. Model prediction results for both the training and prediction sets must be submitted by February 9, 2018. Project results will be presented at a workshop to be held at the National Institutes of Health in Bethesda, Maryland, on April 11-12, 2018.