S3: Navigating through the minefield of read-across
Navigating through the minefield of read-across: from research to practical tools
George Helmana, b, Prachi Pradeepa, b, Imran Shaha
a) National Center for Computational Toxicology (NCCT), Office of Research and Development, US Environmental Protection Agency, 109 TW Alexander Dr, Research Triangle Park (RTP), NC 27711, USA
b) Oak Ridge Institute for Science and Education (ORISE), Oak Ridge, TN, USA.
Read-across is a popular data gap filling technique used within analogue and category approaches for regulatory purposes. In recent years there have been many efforts focused on the challenges involved in read-across development, its scientific justification and documentation. Software tools have also been developed to facilitate read-across development and application. Here, we describe a handful of the publicly available read-across tools in the context of the category/analogue workflow to better articulate their respective capabilities, strengths and weaknesses. Whilst many of these address a number of the steps in the category/analogue workflow, few if any consider uncertainty assessment. To address this gap, we present an algorithmic, automated approach to evaluate the utility of in vitro bioactivity data (“bioactivity descriptors”, from EPA’s ToxCast program) and chemical descriptor information to derive local validity domains (specific sets of nearest neighbors) to enable read-across of toxic effects observed in typical in vivo study types. We demonstrate how performance of a read-across prediction can be evaluated which quantifies the uncertainty. We also showcase the progress that has been made in translating these efforts into practical tools.