期刊论文

中文

化学通报

0441-3776

2023

86

3

312-322

本校为第一机构

随着商品中所含各种化合物的不断使用，人们日益关注其对人类及生态环境的安全危害。在过去的几年里，通过计算方法预测化合物毒性已经显示出极大的潜力。在此，总结了常用的机器学习和深度学习算法在建立毒性预测模型上的优缺点，并系统回顾了近三年发表的可免费访问的毒性预测网络服务器。此外，还讨论了基于人工智能和互联网时代下毒性预测所面临的机遇和挑战。希望指导人们合理选择算法和网络服务器进行建模及化合物毒性评估。

With the continuous using compounds contained in commodities,there is growing concern about their harm to human and ecological environment safety.In the past few years,computational techniques have showed their potential to predict toxicity of compounds.Here,we summariz...MORE With the continuous using compounds contained in commodities,there is growing concern about their harm to human and ecological environment safety.In the past few years,computational techniques have showed their potential to predict toxicity of compounds.Here,we summarize the advantages and drawbacks of machine learning a...