Araport: Baby Steps Towards a Federated Data Portal

Araport has its origins in a series of workshops held about five years ago to assess the needs and future directions of Arabidopsis bioinformatics. For over a decade, The Arabidopsis Information Resource (TAIR) served its research community, primarily by providing high quality, manually curated data pertaining to the model plant Arabidopsis thaliana. However, the exponential growth in relevant data of diverse types and the imminent loss of NSF funding for TAIR led the community to propose a broader and more inclusive model for a next generation data portal. Although the new model proposed that data integration should be driven by locally managed and hosted web services, the reality of the situation was different. Hence, we adopted a hybrid approach where the initial data types and data sets were integrated by a warehousing approach, while at the same time utilizing third party web service wherever available. To this end we instantiated and continue to build on ThaleMine, an Arabidopsis-specific instance of InterMine. We also developed a sophisticated but user-friendly software development kit to allow individual community members to expose their own data through Araport-compliant web services. Over the first two years of the project, we have learned that many community members still perceive a technical barrier to participation. Going forward, our new modus operandi simplifies the interfaces for community user contribution of data or tools, and places the technical workload on the Araport staff. We are optimistic that this will expand both our user base and data content and help us to fulfil the vision of Araport.

Back to: Interoperability and Federation Across Bioinformatic Platforms and Resources

<< Previous Abstract | Next Abstract >>