BioSitemap Reference Implementation Java API

Developers can use biositemaps available on the World-Wide Web to aggregate the information about biomedical resources. Developers can query for resources specified in a biositemap, aggregate resources from different biositemaps and store them in their own RDF database, explore the properties of the resources (such as authors, descriptions, urls, organizations, keywords, etc.), and find resources of specific types. Developers can also use the hierarchy of resource types defined in the BRO to search for resources specifying a more general type. For instance, one can search for all Image_Processing resources and get resources that are tagged with specific Image_Processing resource types, such as Atlas_Generation, Cortical_Modeling, Segementation, and others (all subclasses of Image_Processing in BRO).

This brief user guide presents a Java API for easily discovering, aggregating and querying biositemaps from a Java applications. The BiositemapsConsumer class provides the following functionality:

  • Biositemaps may be located and aggregated using the biositemap registry or read from a local hard disk.
  • Biositemaps may be queried for resources matching an exact or partial string (e.g. wild-card) against any or all properties.
  • Biositemaps may be queried for resources matching a specified BRO or other ontology class.
  • Biositemaps search results may be saved locally.
  • Biositemaps resources may be printed out in a human readable format,

The Javadocs for this reference implementation provide more details.

The jar files are available from the bioistemaps maven repository. The latest version is biostemap-search-api-1.2-beta.jar may be downloaded, and the source code is freely available from subversion:


     $ svn co https://www.umms.med.umich.edu/codestore/ncibi/biositemaps/trunk/api/
	  

The API also comes with a command-line search tool (org.biositemaps.tool.BiositemapsTool) that may be useful for testing and evaluation.


     $ mvn exec:java
	  

Note that building and running the command-line search tool requires Apache Maven.