Despite ongoing harmonization efforts, the major carbohydrate sequence databases following the first initiative in this field, CarbBank, are still isolated islands, with mechanisms for automatic structure exchange and comparison largely missing. This unfavorable situation has been overcome with a systematic data integration effort, resulting in the GlycomeDB, a meta-database for public carbohydrate sequences. It contains at present 35,056 unique structures in GlycoCT encoding, referencing more than 100,000 external records from 1845 different taxonomic sources. We have created a user-friendly, web-based graphical interface which allows taxonomic and structural data to be entered and searched for. The structural search possibilities include substructure search, similarity search, and maximum common substructure. A novel search refinement mechanism allows the assembly of complex queries. With GlycomeDB (www.glycome-db.org), it is now possible to use a single portal to access all digitally encoded, public structural data in glycomics and to perform complex queries with the help of a web-based user interface.