Inhalt

•  
   1 Introduction
•  
  2 Background and related work
•  
  2.1 Principles of data integration
•  
   2.1.1 Link integration and hypertext navigation
•  
   2.1.2 Data warehouses
•  
   2.1.3 View integration and mediator systems
•  
   2.1.4 Workflows
•  
   2.1.5 Mashups
•  
  2.2 Previous work
•  
   2.2.1 SEMEDA
•  
   2.2.2 PROTON
•  
   2.3 Survey of current data integration systems
•  
   2.3.1 Visual Knowledge and BioCAD
•  
   2.3.2 Biozon
•  
   2.3.3 BNDB / BN++
•  
   2.3.4 STRING
•  
   2.3.5 NeAT
•  
   2.4 Conclusion
•  
  3 Requirements
•  
   3.1 Current situation
•  
   3.2 Challenges for data integration
•  
   3.3 Comparison with previous and related work
•  
   3.4 Comparison of approaches to data integration
•  
   3.5 Requirements for ONDEX
•  
   4 Methods and principles
•  
  4.1 ONDEX integration data structure
•  
   4.1.1 Motivation
•  
   4.1.2 Semantics in ONDEX
•  
   4.1.3 Semantics of nodes
•  
   4.1.4 Semantics of edges
•  
   4.1.5 Provenance
•  
   4.1.6 References and synonyms
•  
   4.1.7 Generalised data structure
•  
   4.1.8 Context
•  
   4.1.9 Definition ONDEX integration data structure
•  
   4.1.10 Discussion
•  
  4.2 Data alignment
•  
   4.2.1 Motivation
•  
  4.2.2 Methods
•  
   4.2.2.1 Data import and export
•  
   4.2.2.2 Data integration methods and algorithms
•  
   4.2.2.3 Other data integration methods
•  
   4.2.2.4 Evaluation methods
•  
  4.2.3 Results
•  
   4.2.3.1 Mapping methods – Enzyme Nomenclature vs. Gene Ontology
•  
   4.2.3.2 Mapping methods – KEGG vs. AraCyc
•  
   4.2.3.3 Visualising results
•  
   4.2.4 Discussion
•  
  4.3 Exchanging integrated data
•  
   4.3.1 Motivation
•  
   4.3.2 Requirements for exchanging integrated data sets
•  
  4.3.3 The OXL format
•  
   4.3.3.1 A brief history of OXL
•  
   4.3.3.2 OXL as XML Schema
•  
   4.3.3.3 Support for data integration and text mining
•  
   4.3.3.4 Tool support and applications of OXL
•  
   4.3.3.5 The role of OXL in the ONDEX data integration framework
•  
   4.3.4 Discussion
•  
   4.3.5 Outlook
•  
  5 Design and implementation
•  
   5.1 System design
•  
   5.1.1 Knowledge modelling and domain independence
•  
   5.1.2 Formulating a consensus domain model in biology
•  
   5.1.3 Populating the domain model
•  
   5.1.3.1 Resolving and transforming conflicts
•  
   5.1.3.2 Structured data formats
•  
   5.1.3.3 Dealing with unstructured data
•  
   5.1.4 Data filtering and knowledge extraction
•  
   5.1.5 Workflows
•  
  5.2 Implementing integration data structure
•  
   5.2.1 Encapsulation
•  
   5.2.2 Inheritance
•  
   5.2.3 Association
•  
   5.2.4 Polymorphism
•  
   5.2.5 Aspect oriented development
•  
   5.2.6 Graph implementations and persistency
•  
  5.3 Graph querying and information retrieval
•  
   5.3.1 Views based on semantics
•  
   5.3.2 Indexing and querying
•  
  6 Use cases
•  
  6.1 Improving genome annotations for Arabidopsis thaliana
•  
   6.1.1 Motivation
•  
   6.1.2 Data integration approach
•  
   6.1.3 Data integration exemplar
•  
   6.1.4 Data integration pipeline
•  
   6.1.5 Evaluation of annotation methods
•  
   6.1.6 Discussion
•  
  6.2 Prediction of potential pathogenicity genes in Fusarium graminearum
•  
   6.2.1 Motivation
•  
   6.2.2 Methods
•  
   6.2.3 Results
•  
   6.2.4 Discussion
•  
  6.3 Constructing a consensus metabolic network for Arabidopsis thaliana
•  
   6.3.1 Motivation
•  
   6.3.2 Methods
•  
   6.3.2.1 Graph-pattern mapping
•  
   6.3.2.2 SBML export
•  
   6.3.3 Results
•  
   6.3.4 Discussion
•  
  6.4 Analysis of social networks
•  
   6.4.1 Motivation
•  
   6.4.2 Methods
•  
   6.4.2.1 Friend of a Friend (FOAF) project
•  
   6.4.2.2 Mapping FOAF to the integration data structure
•  
   6.4.3 Results
•  
   6.4.4 Discussion
•  
   6.5 ONDEX SABR project and its applications
•  
   6.5.1.1 Identifying new genetic and molecular targets to improve bioenergy crops
•  
   6.5.1.2 Integration, augmentation and validation of yeast metabolome models
•  
   6.5.1.3 Supporting research into the role of telomere function in ageing
•  
   6.5.1.4 Modelling processes for fruit ripening and flavour development in Tomato
•  
   6.5.1.5 Differences in responses to carcinogenic substances in human, rat and mouse
•  
   6.5.1.6 Plant Responses to Environmental STress in Arabidopsis (PRESTA)
•  
  7 Conclusion and outlook
•  
   7.1 Summary
•  
   7.2 Design decisions
•  
   7.3 Addressing the challenges
•  
   7.4 Comparison with related work
•  
   7.5 Outlook
•  
  8 Glossary
•  
  8.1 Definitions
•  
   8.1.1 Graph theory
•  
   8.1.2 Properties on relations
•  
   8.1.3 Object-oriented development and UML
•  
   8.1.4 Aspect-oriented software development
•  
  8.2 Knowledge representation
•  
   8.2.1 Semantic network
•  
   8.2.2 Conceptual graphs
•  
   8.2.3 Ontology
•  
   8.2.4 Contexts
•  
   8.2.5 Domain knowledge
•  
   8.2.6 Domain modelling
•  
   8.2.7 Hierarchy and taxonomies
•  
   8.2.8 Controlled vocabulary
•  
   8.3 Terms used in ONDEX
•  
  9 Appendix
•  
  9.1 List of data formats supported by ONDEX
•  
   9.1.1 Import
•  
   9.1.2 Export
•  
   9.2 Data integration methods in ONDEX
•  
   Table of figures
•  
   References