Ontologies and Bioinformatics

Overview
Starting with basic definitions and explanations of what Ontologies are all about, the course goes on to cover the technical groundwork of XML semantics, rule-based systems and the semantic web. A survey of the various publicly available Bioinformatics then leads into the main body of the course which covers the details of Building and Using Bioinformatics Ontologies.

Course Benefits
The goal of this course is to provide a firm foundation in the theory and practice of constructing and using Ontologies in general, and in the application of this knowledge to Bioinformatics in particular.

Course Contents
Foundations of Ontologies

• Introduction to XML
  • Markup languages
  • Hierarchical organisation of information
  • Relationships and namespaces
  • Infosets
  • XML Schema
  • XML Data
• Rules and inference
  • Introduction to rule based systems
  • Rule Engine strategies: backward and forward chaining
  • Theorem proving and automated reasoning strategies and algorithms
• Introduction to Semantic Web concepts and applications in Bioinformatics
  • Describing web documents using XML
  • Describing web resources using RDF
  • Understanding the web ontology language OWL
  • Applying logic and inference rules
• Overview of Ontologies in Bioinformatics
  • Bio-Ontologies
    • Unified Medical Language System
    • The Gene Ontology
    • Ontologies of Bioinformatics Ontologies
  • Sequence databases
  • Nucleotide and Protein structure databases
  • Transcription factor databases
  • Species specific databases
  • Specialist protein databases
  • Gene expression databases
  • Pathway databases
  • Single nucleotide polymorphism databases

Construction and Use of Ontologies in Bioinformatics

• Information retrieval
  • Search strategies and mechanisms
  • Vector space information retrieval
  • Ontological approaches to query formulation
  • Organising information via chains of citations
  • Knowledge representation and specialist vs. natural language querying
• Sequence similarity search tools
  • Basic concepts and algorithms
  • BLAST
    • Algorithm
    • Search types
    • Scores and values
    • Variants
• Query languages
  • Navigational - XPath
  • XML querying - XQuery
  • Semantic web queries
• Transformation and data
  • From experimental protocol to results
  • From raw data to processed data
  • Transformations and viewpoints
  • Transformation protocols and technique s
  • Automating the transformation process
• Algorithmic (programming language) transformation techniques (this course will concentrate on Perl as the underlying programming language ... )
  • Text transformation
    • Record oriented transformation
    • Multi-dimensional array transformations
    • Programs and data structures for data transformation
    • Regular expressions and pattern matching
  • Transforming XML
  • Parsing XML: stream-based and tree-based parsing
  • Event Streams
  • SAX
  • Tree processing
  • DOM
  • XPath and XSLT

Building Bioinformatics Ontologies (most of the examples will be based on Protege - any background Java concepts will be explained as needed ... )

• Deciding on the purpose of an Ontology
• Choosing an Ontology language
• Overview of Ontology Development Tools
• Acquisition of domain knowledge
• Reuse of existing Ontologies
• Designing concept hierarchies and properties
• Procedures for validating and revising an Ontology

Handling uncertainty and reasoning with uncertainty

• Inductive vs. deductive reasoning
• Bayesian networks - an introduction
• An Ontology as a Bayesian network
• Validating and revising Bayesian networks
• Combining information
  • Discrete information
  • Continuous information
  • Dempster-Shafer theory
  • The measurement of probability
• Bayesian web and ontologies for Bayesian networks