Search

BACKGROUND: Humans are exposed to tens of thousands of chemical substances that need to be assessed for their potential toxicity. Acute systemic toxicity testing serves as the basis for regulatory hazard classification, labeling, and risk management. However, it is cost- and time-prohibitive to evaluate all new and existing chemicals using traditional rodent acute toxicity tests. In silico models built using existing data facilitate rapid acute toxicity predictions without using animals. OBJECTIVES: The U.S. Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) Acute Toxicity Workgroup organized an international collaboration to develop in silico models for predicting acute oral toxicity based on five different end points: Lethal Dose 50 (LD50 value, U.S. Environmental Protection Agency hazard (four) categories, Globally Harmonized System for Classification and Labeling hazard (five) categories, very toxic chemicals [LD50 (LD50≤50mg/kg)], and nontoxic chemicals (LD50>2,000mg/kg). METHODS: An acute oral toxicity data inventory for 11,992 chemicals was compiled, split into training and evaluation sets, and made available to 35 participating international research groups that submitted a total of 139 predictive models. Predictions that fell within the applicability domains of the submitted models were evaluated using external validation sets. These were then combined into consensus models to leverage strengths of individual approaches. RESULTS: The resulting consensus predictions, which leverage the collective strengths of each individual model, form the Collaborative Acute Toxicity Modeling Suite (CATMoS). CATMoS demonstrated high performance in terms of accuracy and robustness when compared with in vivo results. DISCUSSION: CATMoS is being evaluated by regulatory agencies for its utility and applicability as a potential replacement for in vivo rat acute oral toxicity studies. CATMoS predictions for more than 800,000 chemicals have been made available via the National Toxicology Program's Integrated Chemical Environment tools and data sets (ice.ntp.niehs.nih.gov). The models are also implemented in a free, standalone, open-source tool, OPERA, which allows predictions of new and untested chemicals to be made. https://doi.org/10.1289/EHP8495.

Toxicology data are generated on large scales by toxicogenomic studies and high-throughput screening (HTS) programmes, and on smaller scales by traditional methods. Both big and small data have value for elucidating toxicological mechanisms and pathways that are perturbed by chemical stressors. In addition, years of investigations comprise a wealth of knowledge as reported in the literature that is also used to interpret new data, though knowledge is not often captured in traditional databases. With the big data era, computer automation to analyse and interpret datasets is needed, which requires aggregation of data and knowledge from all available sources. This chapter reviews ongoing efforts to aggregate toxicological knowledge in a knowledge base, based on the Adverse Outcome Pathways framework, and provides examples of data integration and inferential analysis for use in (predictive) toxicology.

The replacement of animal use in testing for regulatory classification of skin sensitizers is a priority for US federal agencies that use data from such testing. Machine learning models that classify substances as sensitizers or non-sensitizers without using animal data have been developed and evaluated. Because some regulatory agencies require that sensitizers be further classified into potency categories, we developed statistical models to predict skin sensitization potency for murine local lymph node assay (LLNA) and human outcomes. Input variables for our models included six physicochemical properties and data from three non-animal test methods: direct peptide reactivity assay; human cell line activation test; and KeratinoSens™ assay. Models were built to predict three potency categories using four machine learning approaches and were validated using external test sets and leave-one-out cross-validation. A one-tiered strategy modeled all three categories of response together while a two-tiered strategy modeled sensitizer/non-sensitizer responses and then classified the sensitizers as strong or weak sensitizers. The two-tiered model using the support vector machine with all assay and physicochemical data inputs provided the best performance, yielding accuracy of 88% for prediction of LLNA outcomes (120 substances) and 81% for prediction of human test outcomes (87 substances). The best one-tiered model predicted LLNA outcomes with 78% accuracy and human outcomes with 75% accuracy. By comparison, the LLNA predicts human potency categories with 69% accuracy (60 of 87 substances correctly categorized). These results suggest that computational models using non-animal methods may provide valuable information for assessing skin sensitization potency. Copyright © 2017 John Wiley & Sons, Ltd.

Newin vitrotesting strategies make it possible to design testing batteries for large numbers of environmental chemicals. Full utilization of the results requires knowledge of the underlying biological networks and the adverse outcome pathways (AOPs) that describe the route from early molecular perturbations to an adverse outcome. Curation of a formal AOP is a time-intensive process and a rate-limiting step to designing these test batteries. Here, we describe a method for integrating publicly available data in order to generate computationally predicted AOP (cpAOP) scaffolds, which can be leveraged by domain experts to shorten the time for formal AOP development. A network-based workflow was used to facilitate the integration of multiple data types to generate cpAOPs. Edges between graph entities were identified through direct experimental or literature information, or computationally inferred using frequent itemset mining. Data from the TG-GATEs and ToxCast programs were used to channel large-scale toxicogenomics information into a cpAOP network (cpAOPnet) of over 20 000 relationships describing connections between chemical treatments, phenotypes, and perturbed pathways as measured by differential gene expression and high-throughput screening targets. The resulting fatty liver cpAOPnet is available as a resource to the community. Subnetworks of cpAOPs for a reference chemical (carbon tetrachloride, CCl4) and outcome (fatty liver) were compared with published mechanistic descriptions. In both cases, the computational approaches approximated the manually curated AOPs. The cpAOPnet can be used for accelerating expert-curated AOP development and to identify pathway targets that lack genomic markers or high-throughput screening tests. It can also facilitate identification of key events for designing test batteries and for classification and grouping of chemicals for follow up testing.

The adverse outcome pathway (AOP) concept links molecular perturbations with organism and population-level outcomes to support high-throughput toxicity (HTT) testing. International efforts are underway to define AOPs and store the information supporting these AOPs in a central knowledge base; however, this process is currently labor-intensive and time-consuming. Publicly available data sources provide a wealth of information that could be used to define computationally predicted AOPs (cpAOPs), which could serve as a basis for creating expert-derived AOPs in a much more efficient way. Computational tools for mining large datasets provide the means for extracting and organizing the information captured in these public data sources. Using cpAOPs as a starting point for expert-derived AOPs should accelerate AOP development. Coupling this with tools to coordinate and facilitate the expert development efforts will increase the number and quality of AOPs produced, which should play a key role in advancing the adoption of HTT testing, thereby reducing the use of animals in toxicity testing and greatly increasing the number of chemicals that can be tested.

SUMMARY: Access to high-quality reference data is essential for the development, validation, and implementation of in vitro and in silico approaches that reduce and replace the use of animals in toxicity testing. Currently, these data must often be pooled from a variety of disparate sources to efficiently link a set of assay responses and model predictions to an outcome or hazard classification. To provide a central access point for these purposes, the National Toxicology Program Interagency Center for the Evaluation of Alternative Toxicological Methods developed the Integrated Chemical Environment (ICE) web resource. The ICE data integrator allows users to retrieve and combine data sets and to develop hypotheses through data exploration. Open-source computational workflows and models will be available for download and application to local data. ICE currently includes curated in vivo test data, reference chemical information, in vitro assay data (including Tox21TM/ToxCast™ high-throughput screening data), and in silico model predictions. Users can query these data collections focusing on end points of interest such as acute systemic toxicity, endocrine disruption, skin sensitization, and many others. ICE is publicly accessible at https://ice.ntp.niehs.nih.gov. https://doi.org/10.1289/EHP1759.

As the number of chemicals and environmental toxicants in commerce continue to increase, so does the need to understand the links between exposure to these stressors and any potential toxic reactions. Assessing the impact of these stressors on public health as well as our environment requires an understanding of the underlying mechanistic processes connecting their introduction into the environment to the associated adverse outcomes. Traditional in vivo methods of toxicity testing have become too costly and inefficient. In recent times, in vitro high-throughput toxicity screening methods have been introduced to reduce the burden of in vivo testing and keep pace with the ever increasing number of required tests. The adverse outcome pathway (AOP) concept has been adopted by many in the toxicology community as a framework for linking the biological events that occur from the point of contact with these stressors and the resulting adverse outcome. This provides a mechanistic framework for understanding the potential impacts of perturbations that are measured via in vitro testing strategies. The aggregate exposure pathway (AEP) has been proposed as a companion framework to the AOP. The goal of the AEP is to describe the path the introduction of the stressor into the environment at its source to a target site within an individual that is comparable with the concentrations in the in vitro toxicity tests. Together, these frameworks provide a comprehensive view of the source to adverse outcome continuum. Standardizing our representation of the mechanistic information in this way allows for increased interoperability for computational models describing different parts of the system. It also aids in translating new research in exposure science and toxicology for risk assessors and decision makers when assessing the impact of specific stressors on endpoints of regulatory significance.

Environmental health risk assessors are challenged to understand and incorporate new data streams as the field of toxicology continues to adopt new molecular and systems biology technologies. Systematic screening reviews can help risk assessors and assessment teams determine which studies to consider for inclusion in a human health assessment. A tool for systematic reviews should be standardized and transparent in order to consistently determine which studies meet minimum quality criteria prior to performing in-depth analyses of the data. The Systematic Omics Analysis Review (SOAR) tool is focused on assisting risk assessment support teams in performing systematic reviews of transcriptomic studies. SOAR is a spreadsheet tool of 35 objective questions developed by domain experts, focused on transcriptomic microarray studies, and including four main topics: test system, test substance, experimental design, and microarray data. The tool will be used as a guide to identify studies that meet basic published quality criteria, such as those defined by the Minimum Information About a Microarray Experiment standard and the Toxicological Data Reliability Assessment Tool. Seven scientists were recruited to test the tool by using it to independently rate 15 published manuscripts that study chemical exposures with microarrays. Using their feedback, questions were weighted based on importance of the information and a suitability cutoff was set for each of the four topic sections. The final validation resulted in 100% agreement between the users on four separate manuscripts, showing that the SOAR tool may be used to facilitate the standardized and transparent screening of microarray literature for environmental human health risk assessment.