Search

Benchmark dose (BMD) modeling is a method commonly used in chemical toxicology to identify the point of departure (POD) from a dose-response curve linked to a health-related outcome. Recently, its application in the analysis of transcriptional data for quantitative adverse outcome pathway (AOP) development is being explored. As AOPs are informed by diverse data types, it is important to understand the impact of study parameters such as dose selection, the number of replicates and dose range on BMD outputs for radiation-induced genes and pathways. Data were selected from the Gene Expression Omnibus (GSE52403) that featured gene expression profiles of peripheral blood samples from C57BL/6 mice 6 hours post-exposure to 137Cs gamma-radiation at 0, 1, 2, 3, 4.5, 6, 8 and 10.5 Gy. The dataset comprised a broad dose range over multiple dose points with consistent dose spacing and multiple biological replicates. This dataset was ideal for systematically transforming across three categories: (1) dose range, (2) dose-spacing and (3) number of controls/replicates. Across these categories, 29 transformed datasets were compared to the original dataset to determine the impact of each transformation on the BMD outputs. Most of the experimental changes did not impact the BMD outputs. The transformed datasets were largely consistent with the original dataset in terms of the number of reproduced genes modeled and absolute BMD values for genes and pathways. Variations in dose selection identified the importance of the absolute value of the lowest and second dose. It was determined that dose selection should include at least two doses <1 Gy and two >5 Gy to achieve meaningful BMD outputs. Changes to the number of biological replicates in the control and non-zero dose groups impacted the overall accuracy and precision of the BMD outputs as well as the ability to fit dose-response models consistent with the original dataset. Successful application of transcriptomic BMD modeling for radiation datasets requires considerations of the exposure dose and the number of biological replicates. Most important is the selection of the lowest doses and dose spacing. Reflections on these parameters in experimental design will provide meaningful BMD outputs that could correlate well to apical endpoints of relevance to radiation exposure assessment. [ABSTRACT FROM AUTHOR]

Transcriptomics data are relevant to address a number of challenges in Toxicogenomics (TGx). After careful planning of exposure conditions and data preprocessing, the TGx data can be used in predictive toxicology, where more advanced modelling techniques are applied. The large volume of molecular profiles produced by omics-based technologies allows the development and application of artificial intelligence (AI) methods in TGx. Indeed, the publicly available omics datasets are constantly increasing together with a plethora of different methods that are made available to facilitate their analysis, interpretation and the generation of accurate and stable predictive models. In this review, we present the state-of-the-art of data modelling applied to transcriptomics data in TGx. We show how the benchmark dose (BMD) analysis can be applied to TGx data. We review read across and adverse outcome pathways (AOP) modelling methodologies. We discuss how network-based approaches can be successfully employed to clarify the mechanism of action (MOA) or specific biomarkers of exposure. We also describe the main AI methodologies applied to TGx data to create predictive classification and regression models and we address current challenges. Finally, we present a short description of deep learning (DL) and data integration methodologies applied in these contexts. Modelling of TGx data represents a valuable tool for more accurate chemical safety assessment. This review is the third part of a three-article series on Transcriptomics in Toxicogenomics.