Associate Professor,
Bioinformatics Center,
Institute for Chemical Research, Kyoto University

Database of Chemical Compounds and Glycans for Pharmacogenomics


     We are developing a database of chemical structures with the information on reactions and pathways, which allows us genome-scale analysis such as expression data analysis. In this study, we mainly focus on the carbohydrate structure database GLYCAN and develop glycan related technologies for the pharmacogenomics applications.


1. Carbohydrate structure database: GLYCAN
GLYCAN is a database of carbohydrate structures collected from research articles. The main portion of GLYCAN came from CarbBank, which is the most famous carbohydrate database but has been unfortunately discontinued due to the lack of funding. We reorganized the structures into newly developed KCF (KEGG Chemical Function) format, which is more suitable for structure-based applications, e.g. inputting new data and searching similar structures.

2. Composite structure map: CSM
Composite structure map is a map illustrating all possible variations of carbohydrate structures within organisms, where the edges and nodes represent glycosyltrasferases and carbohydrate structures (Figure). It is constructed as a bridge between carbohydrate functions and structures, and is able to display, for example, expression data of glycosyltransferases in a compact manner.

3. Carbohydrate structure input tool: KegDraw

KegDraw is an intuitive drawing tool for carbohydrate structures. KegDraw is a Java application, so it runs on locally in a platform-independent manner and it allows the drawing of not only glycan structures but also atomic coordination of chemical compounds. Glycan structures drawn in KegDraw can be used as queries to search against GLYCAN and other databases by KCaM structure search server and as inputs for the database updates.