1. The global prevalence of metabolic disorders, including obesity, is a major concern.
2. High-throughput screening programs such as ToxCast and Tox21 provide activity estimates for thousands of chemicals across hundreds of in vitro assays.
3. This article evaluates the performance of two published ToxPi models (8-Slice vs. 5-Slice) using 38 semivolatile environmental chemicals included in Phase III ToxCast, comparing their performance against previous 3T3-L1 results and literature consensus calls.
The article “On the Utility of ToxCast-Based Predictive Models to Evaluate Potential Metabolic Disruption by Environmental Chemicals” is an informative and well-researched piece that provides a comprehensive overview of the utility of high-throughput screening programs such as ToxCast and Tox21 in evaluating potential metabolic disruption by environmental chemicals. The authors have provided a detailed description of the development timeline for these programs, as well as an evaluation of the predictive success for 38 semivolatile environmental chemicals included in Phase III ToxCast using two published ToxPi models (8-Slice vs. 5-Slice).
The article is generally reliable and trustworthy, with no obvious biases or unsupported claims present. The authors have provided a thorough explanation of the methodology used to evaluate the predictive success for these 38 chemicals, including a comparison against previous 3T3-L1 results and literature consensus calls. Furthermore, they have also validated the performance of this best-performing model by testing a balanced chemical test set of 15 predicted active and 15 predicted inactive chemicals using 3T3-L1 cells.
However, there are some points that could be further explored or considered in more detail in future research on this topic. For example, while the authors have discussed how cytotoxicity filtering can impact predictive success, they do not provide any further details on how this might be done or what methods could be used to achieve this goal. Additionally, while they have discussed how reduction from 8 “slices” or pathways to 5 was achieved through merging certain assays together, it would be beneficial to explore other possible pathways that could be used to reduce slices without sacrificing accuracy or reliability.
In conclusion, overall this article is reliable and trustworthy with no obvious biases or unsupported claims present; however there are some points that could be further explored or considered in more