摘要
近年来以深度学习为代表的人工智能技术与医学、药学等多个领域深度融合。深度学习被应用于蛋白质结构与功能预测、药物靶点预测、药物代谢动力学性质预测、药物有效性及安全性预测以及药物相互作用预测等多个药物研发环节,取得了显著成就,提高研发效率的同时降低临床前试验以及临床试验相关的成本和风险。通过总结多种深度学习方法在药物研发各个过程中的具体应用及分析不同深度学习方法在药物研发中的应用特点,阐述了深度学习在药物研发中现存的一些问题并做出展望,以期为进一步研究提供借鉴的思路和方法。
Recently,artificial intelligence(AI)represented by deep learning(DL)has been deeply integrated with various fields of medical and pharmaceutical sciences.DL has been applied to drug research and development with significant progress in the prediction of protein structure and function,drug target,pharmacokinetic properties,drug safety and efficacy,and drug interactions,with improved efficiency of drug research and development and reduced costs and risks of preclinical and clinical trials.This review summarizes the specific applications of various methods of DL in the whole process of drug research and development,with an analysis of the application characteristics of different DL methods in drug research and development.Finally,some problems and prospects of DL in drug research and development are presented so as to provide reference for ideas and methods in further research.
作者
廖俊
徐洁洁
皮志鹏
窦智扬
尚靖
LIAO Jun;XU Jiejie;PI Zhipeng;DOU Zhiyang;SHANG Jing(School of Science,China Pharmaceutical University,Nanjing 211198,China;School of Traditional Chinese Pharmacy,China Pharmaceutical University,Nanjing 211198,China)
出处
《药学进展》
CAS
2020年第5期387-394,共8页
Progress in Pharmaceutical Sciences
基金
国家自然科学基金资助项目(No.81874331)
双一流创新团队资助项目(No.CPU2018GY19)。
关键词
深度学习
药物研发
新机制和新靶点
药物相互作用
药物不良反应
deep learning
drug research and development
new mechanism and new target
drug-drug interaction
adverse drug reaction
