| Summary: | This thesis addresses the problem of collaboration among experimental biologists and modelers in the study of systems biology by using ontology and Semantic Web Services techniques. Modeling in systems biology is concerned with using experimental information and mathematical methods to build quantitative models across different biological scales. This requires interoperation among various knowledge sources and services. Ontology and Semantic Web Services potentially provide an infrastructure to meet this requirement. In our study, we propose an ontology-centered framework within the Semantic Web infrastructure that aims at standardizing various areas of knowledge involved in the biological modeling processes. In this framework, first we specify an ontology-based meta-model for building biological models. This meta-model supports using shared biological ontologies to annotate biological entities in the models, allows semantic queries and automatic discoveries, enables easy model reuse and composition, and serves as a basis to embed external knowledge. We also develop means of transforming biological data sources and data analysis methods into Web Services. These Web Services can then be composed together to perform parameterization in biological modeling. The knowledge of decision-making and workflow of parameterization processes are then recorded by the semantic descriptions of these Web Services, and embedded in model instances built on our proposed meta-model. We use three cases of biological modeling to evaluate our framework. By examining our ontology-centered framework in practice, we conclude that by using ontology to represent biological models and using Semantic Web Services to standardize knowledge components in modeling processes, greater capabilities of knowledge sharing, reuse and collaboration can be achieved. We also conclude that ontology-based biological models with formal semantics are essential to standardize knowledge in compliance with the Semantic Web vision.
|
|---|
