Lecture notes 18 Jan 2001 Per Kraulis
The usefulness of a database can be increased enormously if it is easy to find entries that satisfy certain search criteria. Some examples of searches that a scientist might want to do:
The databases themselves may contain this information, but some software systems must be used to actually perform this kind of search. There are different ways of designing such systems, and two examples are mentioned here.
The Sequence Retrieval System (SRS) developed by Thure Etzold is a system for integrating heterogenous databases. It is based on premade indexes of the items (words, entries, data fields, text,...) found in a set of documents (database files). Apart from the database files themselves, the indexing procedure requires a grammar (Icarus) that describes what different words in the data files mean, how they are to be indexed, and how they cross-reference to other items in other databases. SRS is a web-oriented system located on a server which is accessed through HTML pages and CGI scripts.
SRS started as an academic project, but it is now a commercial system developed and marketed by LION Bioscience AG. However, academic groups can license the SRS system free of charge and set it up at a server in their own lab.
EBI runs an SRS service which can be used by anyone. It indexes a large number of databases, and it also provides a well-defined web interface which allows programs or web sites to create links that query SRS at EBI.
The Entrez system developed and accessible at the NCBI Entrez site. Similar to some extent to the SRS system, it provides search facilities for a large number of databases, and provides links between them. It provides a well-defined web interface which allows programs or web sites to define links that will query Entrez.
However, it appears that the Entrez system is not available to set up at one's own server. It is purely a system for accessing and searching the databases at NCBI.