[11a-N107-1~11] 23.1 Joint Session N "Informatics"

△ [11a-N107-7] High-precision clustering using molecular similarity evaluation with Digital Annealer

〇Akito Maruo^1,2, Hideyuki Jippo¹ (1.Fujitsu, 2.Hokkaido Univ.)

Keywords:Digital Annealer, molecular similarity, clustering

In searching for compounds, methods focusing on the similarity of molecular structures are often used. We have proposed a method for evaluating similarity by solving the maximum common substructure of two molecules with Digital Annealer as a combinatorial optimization problem. However, there is a problem that the accuracy of clustering is not sufficient. In this study, we verified whether the clustering accuracy of the target property could be improved by using the information of the relative error of the descriptor.