September 1, 2005 (Vol. 25, No. 15)
Strong Points: Very thorough, well organized
Weak Points: None
Speaking of protein structure databases, you’ll have to go a long way to find one more ambitious (and largely successful) than GTOP. Hosted at the National Institute of Genetics in Japan, GTOP is a (sort of) acronym for Genomes TO Protein structures and functions. Structures in the database arise from known, published information, as well as 3-D predictions using motif analysis (PROSITE), family classification (PFAM), transmembrane helix predictions (SOSUI), coiled-coil predictions (Multicoil), and repetitive sequence analysis (RepAlign). The most impressive aspect of the database is the structural information returned from a search. Retrieved records display multiple, rotating 3-D projections of the protein, links to alignments with similar protein sequences, identification of modification sites (glycosylation, for example), and much more.