The SeqHound database system in UML. From the bottom up: the system relies on data provided by the NCBI FTP site and the Gene Ontology resource. It uses the NCBI programming toolkit, the database management system (DBMS) and the bzip compression scheme as programming tools. The database is filled and updated using SeqHound parsers, programming tools and NCBI data as input. The database is searched using the SeqHound query interface which is usable in three forms – as CGI-based web pages, as a local API and as a remote API. All applications (top right) are written using the SeqHound API.
Michalickova et al. BMC Bioinformatics 2002 3:32 doi:10.1186/1471-2105-3-32