TOX-AI : Digitalising toxicological databases using Artificial Intelligence and in silico tools for food safety

TOX-AI：使用人工智能和计算机工具将毒理学数据库数字化以确保食品安全

• 批准号: 2548446
• 金额: --
• 依托单位: King's College London
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2021
• 资助国家: 英国
• 起止时间: 2021 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2548446
• 关键词: TOX; AI; Digitalising; toxicological; databases

The food that we eat contain an increasing diversity of chemicals. Some are added deliberately to enhance productivity or properties of food products, others are contaminants. The aim of this PhD research project is to develop a modelling platform to assess the safety of chemicals present in food. This platform will be developed based on mathematical modelling and explores the use of Artificial Intelligence, harnessing data advances from state-of-the-art databases, such as the OpenFoodTox, first to screen and prioritise substances of higher concern, and then to evaluate in a more detailed way those of higher concern.The PhD project will have full access to a series of models available in VEGA, to predict (1) genotoxicity; (2) carcinogenicity (3) developmental toxicity, (4) reproductive toxicity; (5) endocrine disruption; (6) No Observed Effect Levels (NOAEL): (7) Threshold of Toxicological Concern (TTC); and (8) oral adsorption. The uncertainty of the predictions will be used to rank substance with higher concern, due to higher reliability in the prediction.This PhD project will be carried out in close collaboration with researchers at the Food Standards Agency (FSA), the European Food Standards Authority (EFSA), and the Mario Negri Institute, Milano, Italy. Thus, the PhD student will be well supported by supervisory team with cross-disciplinary expertise and be exposed to an inclusive yet collaborative learning environment.

我们吃的食物中含有越来越多的化学物质。有些是故意添加的，以提高食品的生产率或性能，其他是污染物。这个博士研究项目的目的是开发一个模型平台来评估食品中存在的化学物质的安全性。该平台将在数学建模的基础上开发，并探索人工智能的使用，利用最先进的数据库（如OpenFoodTox）的先进数据，首先筛选和优先考虑高度关注的物质，然后以更详细的方式评估那些高度关注的物质。博士项目将完全访问VEGA中可用的一系列模型，以预测：(1)遗传毒性；(2)致癌性；(3)发育毒性；(4)生殖毒性；(5)内分泌紊乱；(6)未观察到的影响水平（NOAEL）；(7)毒性关注阈值（TTC）；(8)口腔吸附。由于预测的可靠性较高，预测的不确定性将用于对高度关注的物质进行排序。该博士项目将与食品标准局（FSA）、欧洲食品标准局（EFSA）和意大利米兰马里奥内格里研究所的研究人员密切合作。因此，博士生将得到跨学科专业团队的良好支持，并接触到一个包容而合作的学习环境。