2017 Fiscal Year Final Research Report
The application of efficient discrete optimization algorithms to the development of a computer automated method for the elucidation of chemical structures
Project/Area Number |
15K15944
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Multi-year Fund |
Research Field |
Mathematical informatics
|
Research Institution | Nanzan University |
Principal Investigator |
KOICHI Shungo 南山大学, 理工学部, 准教授 (50513602)
|
Project Period (FY) |
2015-04-01 – 2018-03-31
|
Keywords | 離散最適化 / 化合物構造推定 / データベース / 核磁気共鳴分析法 |
Outline of Final Research Achievements |
The aim of this project is to establish a computer automated method for the elucidation of chemical structures using an NMR spectrum-structure database. For this purpose, we used several efficient discrete optimization algorithms to solve some combinatorial problems arising in the development of the method. We first applied an algorithm for the graph inference problem to handle and elucidate highly symmetric structures, which successfully extended the applicability of the method. Next, we addressed a problem in generating orders of retrieved substructures. The generation has to be exhaustive because, for each of the orders, the substructures are merged in the order to form a whole structure, which may depend on the order. This generation problem was resolved with the help of an algorithm for maximum weight spanning tree problem.
|
Free Research Field |
情報化学
|