Abstract
Gclust is a comparative genome database that allows the users to search homologous proteins. It is available at the URL: http://gclust.c.u-tokyo.ac.jp/. Protein data of various different organisms are included, but the database is most useful for plant-related researches. The database is constructed by clustering with the Gclust software of the all-against-all BLASTP data obtained with the total predicted proteins in a set of genomes. On the server, the data can be searched by either a key word, or similarity search, or a phylogenetic profile. We exploited the database for finding chloroplast proteins of endosymbiotic origin.
