Research Of Techniques On Ontology-Based Bioinformation Integration In DrSNPGrid

Because the large amounts of bioinformation datasources burst out on the Web with greate heterogeneities and autonomies as well as the complex relationships between the concepts of bioformation, it is so difficulty to integrate bioformation data only by grammar or by structure, therefore, it's greatly expected to do it by semantics. As studied in semantic Web deeper and deeper, ontology gives information well-formed definitions, and brings it possible to interoperate by semantics, which also brings the hope for semantic integration.In this dissertation work, some researches are proposed for semantic integration of heterogeneous bioinformation based on ontology, which will provide domain experts with convenient information services and greatly help them in finding laws of biodata, so the results of our researches will be useful in the era of post-genome. The main contributions of the dissertation are listed as follows:1. the ontology-based mediator/wrapper framework for bioinformation integration in DrSNPGrid(Disease related SNP Grid): After a full analysis of current systems of bioinformation integration, we propose a three-layer-ontology based mediator/wrapper infrustructure, then we propose domain-ontology-based methods for wrapper construction to encapsulate datasources with heterogeneous structure, which are used in separately for wrapping the structured datasources represented by relational databases and semi-structured datatsources represented by HTML pages; Then, we design a two-stage query rewriting: first rewrite the semantic query into several query strategies by ontology inference and locating the semantic terms, then arrange the accessing order of each datasource inside a query strategy based on mapping ontologies, which encapsulate the accessing of different datasources. After that, we propose an ontology-based twi-integration of middle results. With the design above, we implement the bioinformation integration by semantics and provide a foundation for semantic query.2. ontology management: first we design the resource-register-center. Then we propose some methods for ontology management according to the ontologies'roles during information integration, including the creation and maintenance of the domain ontology, schema ontologies and mapping ontologies.3. ontology-based semantic query: we propose a model of semantic query and analyze the implementation of semantic query from semantic layer, data model layer and grammer layer. Then we describe the whole procedure of implemention of semantic query.