Development of Hocdb (homology-based Clustering Database), A Sequence Classification System For Database Searching

摘要

Recent growth of amino acid sequence databases is remarkable. The increase of the entries provides us enourmous amount of information. At th^ same time, however, the growth of databases decreases the efficiency in prediction of structure and function of proteins by database searching. In other words, redundancy in the database generates the problem. For example, the top 1,000 sequences in the output list of the searching through nr database with PSI-BLAST by using a HIV protease as a query are occupied by close relatives of the query HIV protease. The tertiary structures of the HIV protease and the distant relative, asparatic acid protease, are available now:-