Infer the in Vivo Point of Departure with ToxCast in Vitro Assay Data Using a Robust Learning Approach

• Activity Number: 172 - Risk Prediction and Analysis
• Type: Contributed
• Date/Time: Monday, July 30, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Section on Risk Analysis
• Author(s): Dong Wang*
• Companies: FDA National Center for Toxicological Research (NCTR)
• Keywords: Point of Departure; Robust Statistics; High dimensional regression; Risk assessment; IVIVE; Toxicology

Abstract

The development and application of high throughput in vitro assays is an important development for risk assessment in the 21st century. However, significant challenges exist for statistical approaches of relating in vitro readouts to in vivo findings. We developed a high dimensional robust regression model to infer the in vivo point of departure (POD) with in vitro assay data from ToxCast and Tox21 projects. The in vitro PODs were derived and combined with in vivo PODs from ToxRefDB regarding the rat and mouse liver to build a high dimensional robust regression model. This approach separates the chemicals into a majority, well predicted set; and a minority, outlier set. Salient relationships can then be learned from the data. It was used to demonstrate the predictive power of in vitro PODs for in vitro PODs. The accuracy is comparable with extrapolation between related species (mouse and rat). Chemicals in the outlier set tend to also have more biologically variable characteristics. Incorporating robustness is critical for the performance of the model.