Computational Modeling Core

点击翻译按钮获取中文摘要

Pharmacokinetics (PK) and pharmacodynamics (PD) together define the continuum from toxicant exposure to biological perturbations that can cause adverse outcomes. Toxicology increasingly involves computational approaches that complement laboratory studies and provide a more integrated, quantitative and mechanistic basis for human health risk assessment. The Computational Modeling Core (CMC) will provide a suite of computational capabilities to support both the Biomedical and Environmental Engineering Projects. The CMC will: (i) develop physiologically based PK (PBPK) and PD models of dioxins and dioxin-like compounds, (ii) provide Project-oriented bioinformatic support for high-dimensional omic studies, and (iii) provide cross- disciplinary training in computational toxicology. Model development will be coordinated iteratively with laboratory experiments carried out by the Projects. Prior CMC interactions with the Michigan State University (MSU) Superfund Projects have shown that this iterative approach is efficient for hypothesis generation and evaluation. In Specific Aim 1 (SA1) we will develop PBPK models for 2,3,7,8-tetrachlorodiben-p-dioxin (TCDD) and Superfund site-relevant polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs). The models will include induction of hepatic CYP1A2 as a dioxin-binding protein and liver lipid accumulation for more accurate predictions of free PCDD/F concentrations. The models will support hepatotoxicity studies in mice in Project 3 and PCDD/F bioavailability studies of activated carbon-treated soil in Project 5. Human PBPK models will also be developed to help establish tissue dose equivalency between mice and humans for Project 3 and for extrapolation of in vivo PCDD/F exposure levels based on in vitro assays in Projects 1 and 2. In SA2 we will use bioinformatic tools to parse out aryl hydrocarbon receptor (AHR)-mediated cell state trajectories from single- cell RNA sequencing data in human CD5+ B cells (Project 1) and mouse hepatocytes (Project 3), and apply nonlinear dynamical systems analysis to identify novel biomarkers predictive of onset of AHR-mediated toxicity. We will also identify AHR-perturbed gene regulatory networks to inform pathway modeling in SA3. In SA3, dynamical pathway modeling for Project 1 will address the effects of AHR activation on the PD-1 inhibitory signal transduction pathway in CD5+ B cells. For Project 2, a model of induction of hepatic thyroid hormone metabolism by PCDD/Fs will be developed to support population health risk assessment. For Project 3, pathway modeling will focus on disruption of hepatic lipid metabolism through AHR-mediated transcriptional alteration of key liver enzymes. These models will support rigorous investigation of nonlinear dose-responses and provide a strong foundation for research supporting mechanistically-driven risk assessment. CMC will also support Project 4 to model the vitamin B12 salvaging and de novo synthesis pathways in bacteria dehalogenating PCDD/Fs. In SA4, in collaboration with RETCC, a 3-day course on PBPK modeling and a 5-day course on Computational Systems Toxicology: Modeling and Informatics will be offered in alternate years to SRP and risk assessment communities.