Contact: Ravi Pavagada

Email : ravipr[at]uga.edu

OntoVista


OntoVista is an ontology visualization tool, with unique capabilities related to complex (representationally rich) biological and biochemical ontologies. The most important feature of OntoVista is that it can generate semantically enhanced graphs that capture the cognitive representation of a domain by displaying the entities and relationships in a manner that is most familiar to the domain experts. It has extremely powerful querying and navigation capabilities.


I. Motivation


OntoVista is a visualization application that can visualize these ontologies. There are many powerful visualization tools such as Jambalaya, OwlViz, TGVizTab, RDF Gravity etc., but each of them have their own strengths and weaknesses and none of them provides a complete package for querying, navigation, filtering, or the basic capability of visualizing both OWL and RDF. Most visualization tools provide basic features such as querying, and navigation. Not all of them have sophisticated layouts such as tree layout, spring embedded layout, radial layouts etc. for visualizing ontologies. Most tools only provide basic search capabilities such as class or instance search. We found that Jambalaya, created by the Chisel Group at the University of Victoria, provided excellent layouts and navigation capabilities. It uses visualization techniques such as nested graph view, combined with pan, zoom and fisheye-view for interactive navigation. It provides various layouts such as Radial Layout, Spring Layout, horizontal Tree Layout and vertical Tree Layout, to meet different user-preferences. However, the layouts provided by Jambalaya are not customizable and cannot display views that capture the cognitive representations of the domain as perceived by the domain experts. In Jambalaya, all instances/classes in a ontology can only be set to a particular shape and color. It doesn't allow users to set a specific color and shape for a given class or its instances. Layouts in Jambalaya are designed to capture only the graph structure. In order to create domain specific views, layouts should not only be able to change the shape and color of the nodes, but also be able to adjust the physical locations of nodes based on a conceptual location. This was the main motivation behind the development of our customizable layout. The text based search interface in Jambalaya only allows searching of classes and instances. Users are able to navigate to the selected class or instance after selecting them from the search result. While using Jambalaya extensively for the design of ontologies, we found that the searches provided in Jambalaya are not powerful enough for analysis and navigation of complex ontologies. We found a need for advanced search capabilities such as description search, relationship search, triple search, domain-range search and Semantic search while visualizing complex ontologies such as GlycO. Hence our goal was to design a search interface in OntoVista that implements some of the capabilities of ontology-query languages such as SPARQL without requiring the user to learn their rather complex syntax. Jambalaya provides an arc filter that can be used to filter and un-filter edges from the graph view of the ontology. This arc filter however does not take advantage of the property hierarchy that is available in well structured ontologies. Thus, we felt the need for a semantic arc filter for quick filtering and navigation. Thus, we extended the functionality of Jambalaya through customizable layouts and powerful search capabilities such as triple search, domain-range-search, RDF-description search and relationship search.


II. Project Description


This is an Integrated Technology Resource for Biomedical Glycomics: Technological Research and Development Project IV, a project funded by NIH. OntoVista is an ontology visualization tool, with unique capabilities related to complex (representationally rich) biological and biochemical ontologies such as GlycO.. We describe OntoVista's custom layout feature that uses the information in the ontology as described by the domain experts. The information describing how to visualize contents of an ontology does not belong in the ontology itself. Rather, for this purpose we use layout settings provided by the domain experts to display a view that is meaningful to the users. The custom layout was initially designed to display complex carbohydrate molecules in a way that domain experts are used to; the so-called Cartoonist representation. However, the layout is customizable and can generate domain specific views for the ontology. The view generated is very much dependent on the user's configuration settings. The user can also change the settings to generate views that are of user's interest. We have also refined searching and filtering in Jambalaya. Taking advantage of a hierarchical representation of the relationships (i.e. properties and their sub-properties) in the ontology, a semantic arc filter was developed to help users to quickly visualize the nodes connected through particular sets of relationships such as partonomy/containment, chemical interaction/ reaction etc. Thus, Semantic Arc Filter allows users to show/hide all partonomy relationships such as part_of and its sub-properties or chemical interaction relationships such as interacts_with and its sub-properties at once. We have enhanced the basic search capabilities that are there in Jambalaya by adding advanced searches such as relationship search, description or comment search, domain-range search, triple search and semantic search. Additionally, OntoVista provides a Quick Ontology Browser for easy access of class, sub-classes, properties and sub-properties which can be later used in queries while performing searches.


III. Design


Strengths :

OntoVista provides good querying, navigation and searching capabilities. It provides semantic search to disambiguate instances. It can generate custom layout or images based on the configuration settings.

Limitations :

It is a custom tool built to satisfy the current requirements. It is built as a protege plugin and is restricted to protege API. It has a limitation on the size of the ontology it can load, as any other protege plugin. We would be addressing this limitation in the near future.

Reference: David Goldberg, Mark Sutton-Smith, James Paulson, Anne Dell, Automatic annotation of matrix-assisted laser desorption/ionization N-glycan spectra, Proteomics. 2005 Mar;5(4):865-75


Funding: Bioinformatics of Glycan Expression (one of the four components of the "Integrated Technology Resource for Biomedical Glycomics," appox. $6 million+), National Institute of Health, July 1, 2003 - June 30, 2008.