Carbohydrate Structure Database and Other Glycan Databases as an Important Element of Glycoinformatics

— Carbohydrates are one of the most chemically diverse classes of biomolecules. The amount of accumulated information on carbohydrates is far beyond the level allowing navigation in this data ocean without special tools, which are glycomic databases and prognostic services built on top of these data. Existing databases, focused on solving the particular challenges in glycoscience, are not fully compatible with each other in coverage, data formats, and features served to users. Major problems in the modern glyco-databases include data quality, gaps in coverage, and absence of a widely accepted carbohydrate notation. Most demanded are databases with broad coverage, which can provide a universal dataspace on structures, properties, and functions of carbohydrates, associated with taxonomy and other features of their natural sources. In the framework of the Carbohydrate Structure Database (CSDB) project, we created a database architecture aimed at development of the extensible glycoinformatic portal with continuous maintenance and regular content updates. This architecture was implemented in software free of drawbacks typical for glycomic databases. For the 15 years of existence, CSDB has become the main source of data on glycans of microorganisms, and a platform for multiple carbohydrate-related services. This project includes a global-scale database of natural carbohydrates; among its key features are free access, annual data deposition and updates, search and correction of errors (including those in publications), and regular announcement of new services.