论文部分内容阅读
采用系统编码以一原子为中心逐层扩展的思想,在形式上加以改进,同时用核磁共振中最重要的参数一化学位移值表征碳原子所处的状态,实现了结构的数值化,使化学结构更易被计算机存储和处理,也使判断两结构是否全同(两图同构)的难题简化为比较两者化学位移值之和 CSS(Chemical Shift Sum)是否相等的简单问题。进而针对两类子结构检索各自的特点提出了新的基于广度优先的匹配算法,此算法以特征原子作为初始匹配点能大大减少匹配的盲目性。C~(13)-NMR 是一个已积累了大量知识和规律的领域,碳化学位移值又是碳原子核周围电子云分布情况的直接反映,从这些已有的知识和规律出发正确地获取所需信息将会对计算机辅助有机合成路线设计和分子设计等诸多方面起到积极的推进作用。
The idea of using system code to expand layer by layer with an atom as the center is improved in form. At the same time, the state of carbon atoms is characterized by the chemical shift value, which is the most important parameter in NMR, and the numericalization of the structure is realized The chemical structure is easier to be stored and processed by the computer, and the simple problem of judging whether the two structures are identical (two-image isomorphism) is simplified to comparing whether the sum of the chemical shift values CSS (Chemical Shift Sum) is equal. Then, a new breadth-first matching algorithm is proposed according to the characteristics of the two types of substructures. This algorithm uses the feature atom as the initial matching point and can greatly reduce the blindness of matching. C ~ (13) -NMR is a field that has accumulated a great deal of knowledge and rules. Carbon chemical shift value is also a direct reflection of the distribution of electron clouds around carbon nuclei. From these existing knowledge and rules, The information will play a positive role in many aspects such as computer aided organic synthesis route design and molecular design.