LIS 600 Independent Study
XML Schema Mediation in BioMediator
Dr. Terry Brooks
Summer 2006

Project goals:

• To write a wrapper for new data source
• To gain a clear understanding of how mediated schemas are developed and how BioMediator translates between user queries, the BioMediator ontology and data source schemas.

Overview:

The process of adding a new datasource to Bio-Mediator involves the following steps:

1. Identify the datasource and figure out how to query via URL
2. Write a wrapper that queries the database via URL and returns valid XML that can be consumed by the metawrapper
3. Create relationships in the Source Knowledge Base
4. Write directives that map XML returned by the wrappers to the XML of the mediated schema

During the course of this Independent Study, I wrote two wrappers to query the InterPro protein database. The first wrapper queries by InterPro's internal ID or accession number, while the second wrapper queries by a genetic sequence. The results are returned in XML, which is then passed to the meta-wrappers.

After completing the wrappers, the Bio-Mediator team made relevant entity and relationship connections in the Protege Source Knowledge Base.

These metawrappers use "directives" to translate from the source-specific XML to the XML of Bio-Mediator's mediated schema. Once the results have been translated into Bio-Mediator XML, the data is then displayed.

Details:

• Wrappers

Resources:

• Ontology Development 101: A Guide to Creating Your First Ontology by Natalya F. Noy  and Deborah L. McGuinness. A guide to developing ontologies using Protégé as the sample application.
• BioMediator architecture documentation