基于2D分子指纹的分子相似性方法在虚拟筛选中的应用

Application of 2D molecular fingerprints based on molecular similarity in visual screening

随着计算机技术迅速发展和化合物库分子数量的剧增,分子相似性在药物虚拟筛选中的应用也越来越广泛。描述符是分子相似性方法主要组成部分之一,目前已经有大量的描述符应用于分子相似性,本文总结了2D分子指纹的分类和发展,以及基于2D分子指纹的相似性方法在虚拟筛选中的应用。

Molecular similarity is widely used in visual drug screening and QSAR/QSPR. It is known that the quality of a prediction model normally depends on the molecular descriptors that quantify structural features rele- vant to the properties of interest. Many molecular descriptors have been designed to capture the characteristics of structural information. Among them, molecular fingerprints could provide the most detailed description of molecu- lar structures, as evidenced by the fact that they are routinely used and very efficient in the molecular similarity analysis. This paper reviews the recent progress in the application of 2D molecular fingerprints in visual drug screening.