Suppr超能文献

GroupBuild:一种用于从头药物设计的基于片段的方法。

GroupBuild: a fragment-based method for de novo drug design.

作者信息

Rotstein S H, Murcko M A

机构信息

Vertex Pharmaceuticals Incorporated, Cambridge, Massachusetts 02139-4211.

出版信息

J Med Chem. 1993 Jun 11;36(12):1700-10. doi: 10.1021/jm00064a003.

Abstract

A novel method for de novo drug design, GroupBuild, has been developed to suggest chemically reasonable structures which fill the active sites of enzymes. The proposed molecules provide good steric and electrostatic contact with the enzyme and exist in low-energy conformations. These structures are composed entirely of individual functional groups (also known as "building blocks" or "fragments") which the program chooses from a predefined library. User-selected enzyme seed atom(s) may be used to determine the area(s) in which structure generation begins. Alternatively, GroupBuild may begin with a predocked "inhibitor core" from which fragments are grown. For each new fragment generated by the program, several thousand candidates in a variety of locations and orientations are considered. Each of these candidates is scored based on a standard molecular mechanics potential function. The selected fragment and orientation are chosen from among the highest scoring cases. Tests of the method using HIV protease, FK506 binding protein, and human carbonic anhydrase demonstrate that structures similar to known potent inhibitors may be generated with GroupBuild.

摘要

一种用于从头药物设计的新方法——GroupBuild已经被开发出来,用于提出能够填充酶活性位点的化学合理结构。所提出的分子与酶具有良好的空间和静电接触,并以低能量构象存在。这些结构完全由程序从预定义库中选择的单个官能团(也称为“构建块”或“片段”)组成。用户选择的酶种子原子可用于确定结构生成开始的区域。或者,GroupBuild可以从预对接的“抑制剂核心”开始,从该核心生长片段。对于程序生成的每个新片段,会考虑数千个处于各种位置和方向的候选物。这些候选物中的每一个都基于标准分子力学势函数进行评分。所选片段和方向从得分最高的情况中选择。使用HIV蛋白酶、FK506结合蛋白和人碳酸酐酶对该方法进行的测试表明,使用GroupBuild可以生成与已知强效抑制剂相似的结构。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验