CADD和AIDD的药物化学刍议  

Perspective of CADD and AIDD in medicinal chemistry

作　　者：郭宗儒[1] GUO Zong-ru(Institute of Materia Medica,Chinese Academy of Medical Sciences and Peking Union Medical College,Beijing 100050,China)

机构地区：[1]中国医学科学院、北京协和医学院药物研究所,北京100050

出　　处：《药学学报》2023年第10期2931-2941,共11页Acta Pharmaceutica Sinica

摘　　要：人工智能辅助药物发现(AIDD)是计算机辅助药物发现(CADD)的新版本,现代AI的学习能力极大地提升了CADD的能力。20世纪中叶,QSAR开拓了计算机辅助药物研究。在药物靶标结构信息未知的情况下,用回归的方法揭示具有相似骨架结构的化合物中子结构与活性之间的定量关系,以指导药物分子的设计。经过几十年的发展,CADD技术在基于靶标结构的药物发现(SBDD)、基于片段的药物发现(FBDD)、基于靶标结构的虚拟筛选(SBVS)等方面获得了广泛的应用。而AIDD的重大变革,将显著提升计算机辅助药物设计、虚拟筛选和药物靶标发现的能力。本文从药物化学实践的角度,以实例解读CADD和AIDD的关系。Artificial intelligence-aided drug discovery(AIDD)is a new version of computer-aided drug discovery(CADD).AIDD is featured in significantly promoting the performance of conventional CADD.AI markedly enhances the learning ability of CADD.In the 1960s,CADD was established from conventional QSAR approaches,which mainly used regression approaches to derive substructure-activity relationship for compounds with a common scaffold,and guide drug molecular design,figure out the binding features of drugs,and identify potential drug targets.Since the 1990s,structural biology has provided three-dimensional structures of drug targets,enabling drug discovery based on target structure(SBDD),fragment-based drug discovery(FBDD),and structure-based virtual screening(SBVS)with CADD approaches.In the past 30 years,many first in class(FIC)and best in class(BIC)drugs were discovered with CADD.Now,AIDD will further revolutionize CADD by reducing human interventions and mining big chemical and biological data.It is expected that AIDD will significantly enhance the abilities of CADD,virtual screening and drug target identification.This article tries to provide perspectives of CADD and AIDD in medicinal chemistry with case studies.

关 键 词：计算机辅助药物发现 人工智能辅助药物发现 诺氟沙星 维奈托克 halicin 

分 类 号：R914[医药卫生—药物化学]