Brain Architecture Management System
BAMS is an online resource for neural circuitry, an environment for handling brain data from different species. The information stored and processed by BAMS is collated from the scientific literature.
Because of the complex original schema and the dynamic nature of the system, all the data is split into triples and stored as RDF. RDF is a standard model for data interchange on the Web and it's being adopted by the scientific community for its self-documentation and versatility.
We used SPARQL to access the RDF store. SPARQL is a query language and a protocol for accessing RDF that can easily provide a public read-only interface to the data.
The interface is carefully designed to ensure the good functionality and portability across all major browsers.
The XML document in the "Ontology" section is a serialization of the RDF triples that describe the data. The root tag of the document contains a list of namespace prefixes and their aliases, followed by tags describing unique RDF subjects (in the form of alphanumeric IDs if they are blank nodes, or URIs). Inside each tag describing the subject we use one or more tags to describe the corresponding pairs of predicates and objects that form with the subject an RDF triple. The predicates are URIs and make heavy use of namespace prefixes. This document is dynamically generated so it's always up to date. Its public availability facilitates open collaboration among those interested in this area of neuroscience.
Full text search
The full-text search makes use of the excellent postgresql functionality, making up for the SPARQL drawbacks in this area. The result is a reliable search over the triple store, with result highlighting and ordering by relevance.
With the help of some jQuery plugins we implemented features ranging from collapsible headers to treeview descriptions. Because of the big amount of information that needs to be displayed, the pages show only the most important bits, while secondary resources are hidden initially.