Aswani Unnikrishnan is a Senior bioinformatician at Inotiv and a contractor supporting the National Toxicology Program (NTP) Interagency Center for the Evaluation of Alternative Test Methods (NICEATM) with extensive experience in data analysis and development of bioinformatics tools and methods to evaluate and interpret complex toxicological data. With her background in life sciences and computational biology she consistently strives to make an impact in the field of predictive toxicology through her contributions to projects like EPA’s Toxicity Value Database and NIH NICEATM’s Integrated Chemical Environment (ICE).
OpenTox Summer School 2024
Integrated Chemical Environment (ICE): supporting NAM-based chemical assessments
Abstract
Traditional toxicological assessments have relied on results from in vivo toxicity testing approaches. With the development of new approach methodologies (NAMs), researchers can now integrate in vitro data, in silico predictions, and results from computational tools to help make informed decisions. The Integrated Chemical Environment (ICE, https://ice.ntp.niehs.nih.gov/) provides toxicologically relevant data curated by NICEATM and exploratory tools for interpreting these data. ICE has played a significant role in assisting the development, interpretation, and implementation of NAM-based projects. To best accommodate stakeholder requests and support the execution of NAM-based tools and workflows, ICE is continuously evolving to address growing data needs and incorporate advanced features within its tools.
The ICE Search tool allows integration of data from different models and testing systems to provide summary-level information and bioactivity details for chemicals and mixtures of interest. Recent improvements to the tool are aimed at providing additional interactive visualizations that allows users to see the scope and breakdown of their query. However, developing an overall picture of potential bioactivity for a chemical can be a difficult challenge if there are limited data available for the chemical. ICE’s Chemical Quest tool allows users to explore ICE’s database through chemical identifiers (e.g., SMILES and CASRNs) or 2D renderings, providing information on target chemicals and those with similar structures. These similar chemicals can then be entered into any ICE tool to characterize links between structure, chemical characteristics, or assay information. For instance, the chemical characterization tool allows users to view and compare predicted properties of chemicals through property tables and can provide chemical use information to help inform potential exposure scenarios. The ICE Curve Surfer tool can provide interactive visualizations of concentration response curves for curated high throughput screening (cHTS) data that can help support detailed exploration of ToxCast and Tox21 assays. The ICE cHTS data curation pipeline now includes flags for potential technological interference based on interactions between chemicals and assay technology. Additionally, cHTS assay annotations were updated with OBO Foundry-controlled terms to encompass a broader range of biological and toxicological processes.
The ICE Physiologically Based Pharmacokinetics (PBPK) tool can predict tissue-specific chemical concentration profiles resulting from in vivo doses, while the ICE In Vitro–In Vivo Extrapolation (IVIVE) tool translates in vitro activity concentrations to equivalent in vivo dose estimates. Both these tools have been recently updated to incorporate fetal PBTK model from EPA’S httk v2.2.2 to facilitate maternal and/or fetal compartments comparisons. The IVIVE tool has also been updated to provide exposure predictions alongside its Equivalent administered dose (EAD) output allowing users to compare high- and low-level percentiles to EADs, thereby putting in vitro assays into a real-world context. In this presentation, we provide an introduction to ICE tools and demonstrate how it they can provide user-friendly solutions to navigating complex data and supporting various NAM-based chemical assessments.