NMR Center, N.D. Zelinsky Institute of Organic Chemistry, Moscow
Orientation in a rapidly
growing glycoinformatic space requires adequate databases. This mini-review
covers accumulation and processing of published carbohydrate-related information: structures and their properties, taxonomic and bibliographic annotations, experimental analytical data etc. Problems and solutions specific to development of carbohydrate databases are discussed. Existing databases
are compared each to other using their ideology, functionality, coverage, data
quality and interface as criteria. The review emphasizes the characteristics, which a carbohydrate database must
have to be maintainable, error-free, and able to respond to widespread user
requests. The key points being discussed are the database architecture, cross-project integration, existing glycan
description languages and error control procedures. An attempt to develop an
ideal carbohydrate database is demonstrated by the example of Bacterial Carbohydrate
Structure Database (http://csdb.glycoscience.ru).
