Transplant Integrated Search using Apache Solr

The transPLANT distributed search engine is a tool for locating plant genomic and phenotypic data located in multiple repositories across Europe. Data providers maintain local search indices (using the Apache Solr search technology) that can be independently updated, and queried dynamically from a central server. The indexed information is in essence meta data, describing the objects available to browse and download in each resource, and is provided in accordance with a simple schema that classifies these objects according to common data types. This allows for the provision of an interactive faceted search (available at www.transplantdb.eu), through which users can filter large numbers of initial search results according to data type and provides; in addition, an API is provided.

The advantages of this model are (i) independent update of indices is possible without centralised management (ii) any data provider can add their data to the search through registration of the the URL at which they make compliant indices available, thus supporting the idea of an open consortium model (iii) the resulting facility can be utilised by multiple sites by connecting to the API.

While the model has been developed in the transPLANT project, it is of particular interest to plant phenotype data, which are highly diverse and dispersed in local and institutional repositories. Making these data discoverable is of the highest importance and we are developing the model further with a focus on enhancing the data model to better capture the common meta data associated with these data.