Development of metabolite classification support system based on structural information

Abstract

Secondary metabolites are highly species-specific and play a role for the survival of the producing organism within its natural habitat. Systematization of secondary metabolic pathway is necessary to understand species-specific metabolic pathway and to develop new drugs etc. To attain this, we have developed a classification system of metabolites (called MetClassifier) based on chemical structures and metabolic pathways. Though secondary metabolites with known pathways are very limited, their core chemical structures highly relate to metabolic pathways. So pathways can be predicted on the basis of chemical structures of metabolites. MetClassifier provides result of substructure search, substituent fingerprint, and species fingerprint. To construct a substituent fingerprint, unique substituents are gathered from the metabolites obtained by substructure search. The set of unique substituents is referred to as substituent library. Substituent fingerprints are generated as a bit sequence of 0 and 1 based on matching of substituents included in a metabolite with the substituent library. In the same way, species fingerprints are generated. Diversity of substituents can be systematized by substituent fingerprints. This information is useful for elucidating secondary metabolic pathways from a viewpoint of structural similarity of compounds. Relationship between species and substituents can also be systematized by species fingerprints.