Table 1.
Description of data sources used.
| Data stream | Source | Version | Notes |
|---|---|---|---|
| Functional Use Categories | EPA’s Aggregated Computational Toxicology Online Resource (ACToR) | 2014 | Broad use categories (Dionisio et al., 2015; Wambaugh et al., 2014) used in ExpoCast SEEM2 were also used to describe the functional diversity of the 448 substances in this case study. |
| High-throughput bioactivity data | ToxCast | Invitrodb_v3 | This is the public release of invitrodb dated September 2018 (EPA, 2018). These data were fit using the ToxCast Data Pipeline approach (tcpl R package v2). The data used in this case study are available as Supplemental File X. |
| In vitro phenotypic profiles of lung, kidney, and liver cell models (HIPPTox) | Performed by A*STAR for this case study | The cell models and phenotypic readouts were described previously (Lee et al., 2018; Su et al., 2016). All phenotypic readouts (not limited to those predictive of tissue-specific adversary effects) were used in computation of the HIPPTox-POD. | |
| Toxicokinetics | High-throughput toxicokinetic (httk) data | Httk R package v1.8 | Httk R package v1.8 is available from CRAN (https://cran.r-project.org/web/packages/httk/index.html) |
| In vivo PODsa | ToxValDB in vivo toxicity information | Development v5 (May 2018) | This database includes summary point-of-departure information from multiple databases (as described in text) and study types, and is public in the CompTox Chemicals Dashboard. |
| ECHA | Repeated dose study results via the oral route in REACH registration dossiers | These data are publicly available at https://echa.europa.eu/ | |
| EFSA | Published human health risk assessments in support of EU food law 158/2002 | These data include PODs from multiple study types, mostly from acute, subchronic, chronic, and reproduction toxicity studies. | |
| Health Canada | Published risk assessments conducted for existing substances under the Canadian Environmental Protection Act, 1999 | Information was retrieved based on the availability of a published risk assessment conducted under various phases of Canada’s Chemicals Management Plan and earlier initiatives as well as corresponding availability of ToxCast and HTTK data. Point of departure information was extracted from oral repeat-dose studies (of various durations) as well as from developmental and reproductive toxicity studies cited within the assessments. Where possible, both the NO(A)EL and LO(A)EL for each study were collected and the basis for the effect level is described (ECCC/HC, 2016). | |
| Exposure | ExpoCast predictions | Systematic Empirical Evaluation of Models version 2 (SEEM2) | The median and 95th percentile on the credible interval for the total US population exposure estimates were used (Wambaugh, et al., 2014). |
| Health Canada | Published risk assessments conducted for existing substances under the Canadian Environmental Protection Act, 1999 | Exposure estimates were extracted from the same assessments as their respective in vivo POD values. This included the estimated daily intakes from environmental media as well as intakes from use of certain sentinel consumer products (ECCC/HC, 2016). |
All in vivo POD data from source databases were concatenated and are available in Supplemental File 1.
References for Table 1.
Dionisio, K. L., Frame, A. M., Goldsmith, M. R., Wambaugh, J. F., Liddell, A., Cathey, T., Smith, D., Vail, J., Ernstoff, A. S., Fantke, P., et al. (2015). Exploring consumer exposure pathways and patterns of use for chemicals in the environment. Toxicol Rep 2, 228–237.
ECCC/HC (2016). Chemicals Management Plan. In (H. C. Environment and Climate Change, Ed.), Vol. https://www.canada.ca/en/health-canada/services/chemical-substances/chemicals-management-plan.html. Government of Canada, Ottawa (ON).
EPA, U. (2018). ToxCast & Tox21 Data from invitrodb_v3. In (Retrieved from http://www2.epa.gov/chemical-research/toxicity-forecaster-toxcasttm-data.
Lee, J. J., Miller, J. A., Basu, S., Kee, T. V., and Loo, L. H. (2018). Building predictive in vitro pulmonary toxicity assays using high-throughput imaging and artificial intelligence. Arch Toxicol 92(6), 2055–2075.
Su, R., Xiong, S., Zink, D., and Loo, L. H. (2016). High-throughput imaging-based nephrotoxicity prediction for xenobiotics with diverse chemical structures. Arch Toxicol 90(11), 2793–2808.
Wambaugh, J. F., Wang, A., Dionisio, K. L., Frame, A., Egeghy, P., Judson, R., and Setzer, R. W. (2014). High throughput heuristics for prioritizing human exposure to environmental chemicals. Environ Sci Technol 48(21), 12760–7.