Authoring bio-ontologies is a task that traditionally requires contributions from both a domain expert and an ontology engineer. That many domain experts are not also experts in ontology design or in languages as the Web Ontology Language (OWL) is a significant bottleneck in the development process, especially as requests typically outnumber the bandwidth of a single ontology developer, which impacts both ontology development and the ability to semantically describe data typically seen in the Semantic Web. We present the Webulous architecture, a server-client application for developing ontologies based on customisable client interfaces. We also describe the Webulous Google App, a client to Webulous that allows collaborative, online editing of Google Spreadsheets. These spreadsheets can be pre-populated with parts of ontologies loaded from resources such as BioPortal and can be rapidly turned into new ontology terms using behind the scene templates which automatically axiomatise cells in a row to create rich and consistent ontology classes.
Like most data resources, ontologies are rarely complete, and healthy ontologies are continually growing as the state of knowledge progresses. One of the ways in which ontologies grow is through the submission of external requests from users of ontologies. Typically, authoring these ontologies is a task performed by trained experts, familiar with ontology development practices and the complexities of languages such as OWL. OWL ontologies can be developed with dedicated ontology authoring tools such as Protégé1. This can be a barrier to a domain expert in contributing to ontologies and additionally represents a major bottleneck to ontology authoring. In addition, ontologies which represent a particular domain are often seen as consensus models. Such consensus is ideally formed by a number of domain experts collaboratively constructing an ontology. Tools such as Web Protégé2 allow ontology experts to collaboratively develop online, however, this again represents a barrier to a nonontology expert in contributing to content. A solution which allows construction of ontologies alongside the direct contribution of numerous experts is required.
We have developed the Webulous framework and the accompanying Webulous Google App to tackle some of the issues associated with collaborative editing and engaging domain experts who are not ontology developers. Webulous has a server backend that acts as a hub for serving up template spreadsheets, receiving content once a spreadsheet is submitted and for creating new ontology classes or enriching existing classes for addition to an ontology.
The Webulous Google App is a spreadsheet Add On which can enable ontology editing using a Google Spreadsheet online and collaboratively with others. Cells in the spreadsheet can be prepopulated with suggestions from parts of an existing ontology to help editors, for instance all subclasses of the class ‘cancer’ for a column called ‘cancer type’. Once a spreadsheet is complete, it can be submitted and programmatically turned into new classes using a server side design pattern. This create axioms based on the column, for the instance above column B ‘cancer type’ which may be axiomatised as ‘has_disease some <column_B>’ in OWL where column_B is the class entered into a cell.
Webulous has the following primary features: ─ Connect to the Webulous server using Webulous Google Spreadsheet Add On ─ Share and collaborate on the spreadsheet with others ─ Can restrict columns of the Webulous Add On to particular parts of an existing ontology (e.g. all subclasses of disease or human anatomical regions) ─ Search BioPortal for a class and add into the spreadsheet as you create a term ─ Create custom client using the Webulous API 2.1
Availability
The Webulous Service for submitting EFO terms is available at http://www.ebi.ac.uk/efo/webulous The Webulous Google App is available at http:// https://chrome.google.com/webstore/detail/webulous/noieiladpjihajkdgipcmnjcjcgplo od and the Webulous source code for installing your own service is available at https://github.com/EBISPOT/webulous 3
We acknowledge funds from EMBL and The National Center for Biomedical Ontology, one of the National Centers for Biomedical Computing funded by NHGRI, the NHLBI, and the NIH Common Fund under grant U54-HG004028. The CORBEL project is funded by the European Commission within the Horizon 2020 Framework Programme, grant agreement number 654248.