The use of in silico and in vitro methods, commonly referred to as New Approach Methodologies (NAMs), has been proposed to support environmental (and human) chemical safety decisions, ensuring enhanced environmental protection. Toxicokinetic models developed for environmentally relevant species are fundamental to the deployment of a NAMs-based safety strategy, enabling the conversion between external and internal chemical concentrations, although they require historical toxicokinetic data and robust physical models to be considered a viable solution. Daphnia magna is a key model organism in ecotoxicology albeit with limited and scattered quantitative toxicokinetic data, as for most invertebrates, resulting in a lack of robust toxicokinetic models. Moreover, current D. magna models are chemical specific, which restricts their applicability domain. One aim of this study was to address the current data availability limitations by collecting toxicokinetic time-course data for D. magna covering a broad chemical space and assessing the dataset's uniqueness. The collated toxicokinetic dataset included 48 time-courses for 30 chemicals from 17 studies, which was developed into an R package named AquaTK, with 11 studies unique to our work when compared to existing databases. Subsequently, a proof-of-concept Bayesian analysis was developed to estimate the steady-state concentration ratio (internal concentration / external concentration) from the data at three different levels of precision given three different data availability scenarios for environmental risk assessment. Specifically, an atrazine case study illustrates the multi-level modelling approach providing improvements (uncertainty reductions) in predictions of ratios for increasing amounts of data availability. Our work provides a consistent and self-contained Bayesian framework that irrespective of the hierarchy or resolution of individual experiments, can utilise the available information to generate optimal predictions of steady-state concentration ratios in D. magna. This approach is paramount to supporting the implementation of a NAMs based environmental safety paradigm shift in environmental risk assessment.

中文翻译：

---

一种概念验证多层贝叶斯方法，用于整合大型水蚤的理化性质和毒代动力学时程数据


已提议使用计算机和体外方法，通常称为新方法（NAM），以支持环境（和人类）化学安全决策，确保加强环境保护。为环境相关物种开发的毒代动力学模型是部署基于 NAMs 的安全策略的基础，能够实现外部和内部化学浓度之间的转换，尽管它们需要历史毒代动力学数据和强大的物理模型才能被视为可行的解决方案。大型水蚤是生态毒理学中的关键模式生物，尽管与大多数无脊椎动物一样，定量毒代动力学数据有限且分散，导致缺乏强大的毒代动力学模型。此外，目前的 D. magna 模型是化学特异性的，这限制了它们的适用性领域。本研究的目的之一是通过收集涵盖广泛化学空间的 D. magna 的毒代动力学时程数据并评估数据集的独特性来解决当前数据可用性的限制。整理的毒代动力学数据集包括来自 17 项研究的 30 种化学物质的 48 个时间过程，这些时间过程被开发成一个名为 AquaTK 的 R 包，与现有数据库相比，其中 11 项研究是我们工作独有的。随后，开发了一种概念验证贝叶斯分析，以在给定三种不同的数据可用性情景下，以三种不同的精度水平从数据中估计稳态浓度比（内部浓度/外部浓度），以进行环境风险评估。具体来说，阿特拉津案例研究说明了多级建模方法，该方法为增加数据可用性的比率预测提供了改进（降低不确定性）。 我们的工作提供了一个一致且自包含的贝叶斯框架，无论单个实验的层次结构或分辨率如何，都可以利用可用信息来生成 D. magna 稳态浓度比的最佳预测。这种方法对于支持在环境风险评估中实施基于 NAMs 的环境安全范式转变至关重要。