物理化学学报 >> 2012, Vol. 28 >> Issue (03): 536-540.doi: 10.3866/PKU.WHXB201201041

理论与计算化学 上一篇    下一篇

一种在无机晶体结构中检索特定子结构的计算机方法

霍卫峰1, 李乙1, 卢君然1, 于吉红1, 徐如人1, 李晶2   

  1. 1. 吉林大学无机合成与制备化学国家重点实验室, 长春 130012;
    2. 北京科技大学应用力学系, 北京 100083
  • 收稿日期:2011-11-17 修回日期:2011-12-27 发布日期:2012-02-23
  • 通讯作者: 李乙 E-mail:yili@jlu.edu.cn
  • 基金资助:

    国家自然科学基金(21001049)资助项目

A Computational Method for Specified Substructure Search in Inorganic Crystal Structures

HUO Wei-Feng1, LI Yi1, LU Jun-Ran1, YU Ji-Hong1, XU Ru-Ren1, LI Jing2   

  1. 1. State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, Jilin University, Changchun 130012, P. R. China;
    2. Department of Applied Mechanics, University of Science and Technology Beijing, Beijing 100083, P. R. China
  • Received:2011-11-17 Revised:2011-12-27 Published:2012-02-23
  • Contact: LI Yi E-mail:yili@jlu.edu.cn
  • Supported by:

    The project was supported by the National Natural Science Foundation of China (21001049).

摘要: 提出了一种针对无机晶体化合物的子结构检索方法. 该方法以VF2 子图同构算法为基础, 针对无机晶体化合物的结构特点, 采用了两种策略以提高子结构检索的效率: (1) 引入晶体的对称性信息避免了在等价原子间进行的大量重复性计算; (2) 采用结构编码预筛选可以有效地减少目标结构的数量. 我们以在无机微孔分子筛数据库中进行子结构检索为例测试该方法的有效性. 测试结果表明, 该方法可以快速且准确地在分子筛数据库中检索包含特定子结构的记录. 两种检索策略的引入大大降低了子结构检索的复杂度, 检索速度可提高3-5个数量级. 该方法通过Perl 语言实现, 具有较好的可移植性.

关键词: 无机晶体材料, 数据库, 子结构检索, VF2算法, 分子筛, 构筑基元

Abstract: In this paper, a computational method for the substructure search in inorganic crystal structures is proposed. This method is based on the VF2 subgraph isomorphism algorithm. Furthermore, two additional approaches have been introduced into this method to improve the calculation efficiency of VF2: (1) introduction of crystal symmetry information with a view to avoiding redundant calculations among equivalent nodes (atoms); (2) a prescreening encoding treatment to enhance the calculation efficiency by greatly reducing the number of target structures. We tested the efficiency of this method by searching the zeolite crystal structure database from the International Zeolite Association for entries containing specified building units. The test results showed that this method could quickly and correctly retrieve all the entries containing the queried substructure in the zeolite structure database. The introduction of crystal symmetry information and the prescreening encoding treatment greatly reduce the complexity of substructure search. The search speed was significantly enhanced by at least 3-5 orders of magnitude. This method was developed using Perl programming language, ensuring that this method could be easily applied to various platforms.

Key words: Inorganic crystal structure, Database, Substructure search, VF2 algorithm, Zeolite, Building unit

MSC2000: 

  • O641