You are here:
Reproducibility and Reliability of High-Throughput Transcriptomics for Chemical Safety Screening
Citation:
Everett, L., D. Haggard, J. Bundy, B. Vallanat, J. Harrill, R. Judson, AND I. Shah. Reproducibility and Reliability of High-Throughput Transcriptomics for Chemical Safety Screening. Society of Toxicology 2021 Annual Meeting (Virtual Event), Virtual, NC, March 12 - 26, 2021. https://doi.org/10.23645/epacomptox.18130784
Impact/Purpose:
Poster presented at the Society of Toxicology 2021 annual meeting in March 2021. High-Throughput Transcriptomics (HTTr) is a promising method for safety screening and prioritization of environmental chemicals. Decreasing costs have made it feasible to profile all protein-coding genes across thousands of samples, allowing for broad evaluation of many target pathways and modes of action in a single screening assay. US EPA is developing HTTr methods to rapidly screen chemicals in vitro, and has currently analyzed over 1,000 chemicals in 3 biologically distinct cell culture systems. The resulting data can be used for both hazard prediction and potency estimation, thereby informing risk assessments and prioritizing chemicals for further study. Both the reliability and reproducibility of this screening platform are critical to the utility of HTTr in regulatory applications. In this work, we characterize the reproducibility and reliability of HTTr using reference samples and chemicals included systematically throughout these large-scale screening studies.
Description:
High-Throughput Transcriptomics (HTTr) is a promising method for safety screening and prioritization of environmental chemicals. Decreasing costs have made it feasible to profile all protein-coding genes across thousands of samples, allowing for broad evaluation of many target pathways and modes of action in a single screening assay. US EPA is developing HTTr methods to rapidly screen chemicals in vitro, and has currently analyzed over 1,000 chemicals in 3 biologically distinct cell culture systems. The resulting data can be used for both hazard prediction and potency estimation, thereby informing risk assessments and prioritizing chemicals for further study. Both the reliability and reproducibility of this screening platform are critical to the utility of HTTr in regulatory applications. In this work, we characterize the reproducibility and reliability of HTTr using reference samples and chemicals included systematically throughout these large-scale screening studies. First, we demonstrate the reliability of the HTTr assay using 4 standard reference samples that were profiled in duplicate on each of 333 individual screening plates across four independent studies utilizing several variants of the TempO-seq platform. We then demonstrate the high reproducibility of our complete HTTr chemical screening workflow—which also utilizes acoustic liquid handling systems to rapidly automate chemical treatments—by profiling reference chemicals that were replicated on each screening plate. Based on these reference treatments, we characterize the reproducibility of transcriptomic perturbation observed at multiple levels, from individual probes to gene sets capturing the activity of known biological pathways and signatures. We also demonstrate the high correlation across replicates for both single concentration response profiles and for biological pathway altering concentrations (BPACs) inferred from multi-concentration curve-fitting models. Overall, our results demonstrate that the high reproducibility and reliability of our workflow is sufficient to inform risk assessment and prioritization of chemicals in a tiered-testing strategy. These results also inform best practices for HTTr data analysis, including sample-level quality control, and optimal methods for concentration-response modeling. This abstract does not necessarily reflect US EPA policy.
URLs/Downloads:
DOI: Reproducibility and Reliability of High-Throughput Transcriptomics for Chemical Safety Screening
EVERETT_SOT_2021_HTTR_REPRODUCABILITY_V4.PDF (PDF, NA pp, 1009.666 KB, about PDF)