Abstract
Simple and complex carbohydrates have been described as "the last frontier of molecular and cell biology". The enzymes that are required for the synthesis and degradation of these compounds provide an enormous challenge in the post-genomic era. This reflects both the extreme chemical and functional diversity of sugars and the difficulties in characterizing both the substrates and the enzymes themselves. The vast myriad of enzymes involved in the synthesis, modification and degradation of oligosaccharides and polysaccharides is only just being unveiled by genomic sequencing. These so-called "carbohydrate-active enzymes" lend themselves to classification by sensitive sequence similarity detection methods. The modularity, often extremely complex, of these enzymes must first be dissected and annotated before high throughput characterization or "structural genomics" approaches may be employed. Once achieved, modular analysis also permits collation of a detailed "census" of carbohydrate-active enzymes for a whole organism or throughout an ecosystem. At the structural level, improvements in X-ray crystallography have opened up a three-dimensional understanding of the way these enzymes work. The mechanisms of many of the glycoside hydrolase families are becoming clearer, yet glycosyltransferases are only slowly revealing their secrets. What is clear from the genomic and structural data is that if we are to harness the latent power of glycogenomics, scientists must consider distant sequence relatives revealed by the sequence families or other sensitive detection methods.