基于GSCCT的族性化学结构筛选方案

Strategy for screening generic chemical structures based on GSCCT representation

以处理族性结构信息的计算机表达式—族性结构紧缩关联表(Generic Structure Compact Connection Table,GSCCT)为基础,拟定了一套检索族性结构的筛选策略,即从GSCCT 表中提取出主干环节点的预筛选方案。GSCCT 表包含主干结构节点和叶结构节点,主干节点又分为环节点和非环节点两部分。叶结构节点中含有环节点时,将其提升为主干环节点。该结构匹配方法与传统的在原子节点层次上的算法不同,是在紧缩节点的层次上提取关键信息,即提取族性结构中的主要信息-环结构信息(或称指纹信息)进行预筛选,先不考虑非环节点和叶节点,以避免大量枚举。文中详细介绍了筛选思路和筛选功能的实现过程。

Based on the representation of generic chemical structure called Generic Structure Compact Connection Table(GSCCT),astrategy is presented for screening generic chemical structure that is screening the stem ring node from GSCCT.The GSCCT is com-posed of a stem node table and a leaf node table.A stem node may be a ring node or a non-ring node.A leaf node sometime containsring nodes,we extract the ring nodes from the leaf node table to the stem node table.The screening approach is conducted on the com-pact node level,which is different from the traditional search method of atom-by-atom.In order to avoid enumeration of the com-pounds,the ring structure information or finger print information is used as screens.Non-ring and leaf nodes are not considered.Theprinciple of the screening process and its implementation are described in detail.