人类健康
计算机科学
数量结构-活动关系
风险分析(工程)
风险评估
生化工程
标准化
适用范围
危害
危害分析
化妆品
透明度(行为)
领域(数学分析)
数据科学
药物开发
计算模型
预测建模
管理科学
毒物动力学
作者
Pauline Lancia,Myriam Louazzani,Ludivine Gros,José Ginestar,Elena Fioravanzo,Aurélie Baleydier
标识
DOI:10.1021/acs.chemrestox.4c00534
摘要
The development of alternative methods to animal testing has gained momentum over the years, including the rapid growth of in silico methods, which are faster and more cost-effective. A large number of computational tools have been published, focusing on Read-Across, (quantitative) Structure–Activity Relationship ((Q)SAR) models, and Physiologically Based Pharmacokinetic (PBPK) models. All of these methods play a crucial role in the risk assessment for cosmetics. However, despite the continuous efforts of various working groups, these methods are not always accepted by regulatory authorities around the world due to a lack of standardization and transparency in their development and application. This study aimed to identify in silico tools that can predict key properties relevant to the hazard assessment of cosmetic ingredients, aiming to streamline decision-making and assist toxicologists in efficiently selecting and integrating in silico predictions. Eighty-four in silico tools were identified based on their predictive capabilities, covering physicochemical parameters, toxicological/ecotoxicological endpoints, and toxicokinetic properties using different computational methods, e.g., (Q)SARs; Read-Across. Additional criteria were also considered for QSAR models, helping toxicologists integrate them into risk assessment processes: (1) definition of the Applicability Domain (AD), (2) model performance, and (3) nearest neighbors of the target substance. Based on these criteria, the models were classified as either useful for screening or suitable for a Weight of Evidence (WoE) approach. Finally, this study highlights the growing number of computational tools available for assessing various endpoints relevant to cosmetic safety. The number of tools continues to increase, and regular reviews are necessary. A deeper understanding of these in silico tools will facilitate their use by toxicologists and improve their acceptance for regulatory purposes from different cosmetic authorities.
科研通智能强力驱动
Strongly Powered by AbleSci AI