Three-dimensional structural similarity search of molecules based on Hetero-tree representation

Abstract

It is well known that molecular properties, including biological functions, are closely related not only to the atomic connectivity but also to the 3D spatial arrangements of the atoms. In the present work, the hetero-atom filter reduced representation of a molecular structure was employed. Structural information of a molecule including its 3D geometry was represented by a weighted graph of which the nodes and edges correspond to hetero-atoms and the inter-atomic distances between them, respectively. To describe local 3D structural information of a molecule, the authors have defined the Hetero-tree representation, i.e., the minimum spanning tree (MST) of such molecular graph based on inter-atomic distances between hetero-atoms.We have prepared a target data set that contains 107 ester molecular structures taken from CSD. Each molecule of the data set contains from six to ten hetero-atoms, and was represented by its Hetero-tree. Structural feature searching was carried out based on the topological information of Hetero-trees. For example, there were 23 molecules that have the common Hetero tree, as n-hexane type. As a result, we successfully found the molecules that have the common structural feature based on Hetero-tree but quite different scaffold structures.