1. Mining the Biomedical Research Literature Ken Baclawski

2. Data Formats n Flat files n Spreadsheets n Relational databases n Web sites

3. XML Documents n Flexible very popular text format n Self-describing records

4. XML Documents (continued) n Hierarchical structure

5. Ontologies n An ontology defines the concepts and relationships between them in a domain. n Philosophers speak of “the” ontology and define it informally. In Computer Science there are many ontologies and they are formally defined. n The structure of data is its ontology. – Database schema – XML Document Type Definition (DTD)

6. Gene Regulation Ontology Gene Protein Upstream Region Species Motif Protein Domain Entity Regulatory Entity Regulation Regulatory Region Genomic Location Fuzzy DNA Motif Algorithm

7. Gene Protein Species Motif Protein Domain domainPosition domainLength Entity name genbankID Regulatory Entity genomicLength genomicDirection Regulation conformation regulatoryStrength regulatoryType Regulatory Region regionStartPosition regionEndPosition posteriorProbabililty Genomic Location region position unit Fuzzy DNA Motif Algorithm species location activated by sequence contained in regulates motif source

8. Constructing Large Ontologies Base Scenario Event 5 Event 4 Event 3 Event 2 Event 1 Boolean Logic Fuzzy Logic Component Ontology Component Ontology Component Ontology Domain Specific Ontology Domain Specific and Logic Specific Ontology Boolean Logic Probabilistic Logic

9. Some Ontology Languages n Established languages – Knowledge Interchange Format (KIF) – XML Schema (XSD) – Resource Description Framework (RDF) – XML Topic Maps (XTM) n Emerging languages – Common Logic – Web Ontology Languages (OWL) – Ontology Definition Metamodel (ODM)

10. Biomedical Ontologies n Gene Ontology (GO) n Unified Medical Language System (UMLS) n BioPolymer Markup Language (BioML) n Systems Biology ML (SBML) n MicroArray Gene Expression ML (MAGE-ML) n Protein XML (PROXIML) n CellML n RNAML n Chemical ML (CML) n Medical ML (MML) n CytometryML n Taxonomic ML (TML)

11. n Semantic Categories – > 130 semantic categories n Semantic Relationships – “ is a “, “ part of”, “disrupts” n Semantic Concepts (Vocabulary) – > 1,000,000 concepts map to categories

12. Natural Language Processing using an Ontology syntactic semantic

13. Example of knowledge extraction CAMPATH-1 antibodies recognize the CD52 antigen which is a small lipid-anchored glycoprotein abundantly expressed on T cells, B cells, monocytes and macrophages. They lyse lymphocytes... CAMPATH-1 antibody CD52 antigen lymphocyteslysis Is a Interacts with causes affects glycoprotein Is a

14. Purpose of Data n Data is collected and stored for a purpose. n The format serves that purpose. n Using data for another purpose is common. n It is important to anticipate that data will be used for many purposes. n Data is reused by transforming it.

15. Statistical Analysis n Transformation consists of a series of steps. n Specialized equipment and software is used for each step. n Separation into steps reduces the overall effort.

16. Statistical Models n The “selection” step can involve much more than just choosing fields of a record: – Data can be rescaled, discretized,... – Data in several fields can be combined. – The statistical model can be much more complex (such as a Bayesian network). n In general, data is transformed to a different ontology: A statistical model is an ontology.

17. Transformation Languages n Traditional programming languages such as Perl, Java, etc. n Rule-based (declarative) languages such as the XML Transformation language (XSLT). – Rule-based rather than procedural – Transform each kind of element with a template – Matching and processing of elements is analogous to the digestion of polymers with enzymes.

18. High Performance Indexing Document Knowledge Representation NLP fragmentation Knowledge FragmentsDistributed Index Engine Query NLP Knowledge Representation fragmentation Knowledge Fragments Matching Documents

19. Consistency Checking n Logical consistency means that a formal theory has at least one interpretation. n Inconsistency is to be avoided. n Probabilistic consistency means that a probabilistic model is likely to have an interpretation. n Probabilistic inconsistency is significant.

20. Research Challenges n Inference and deduction – Logical inference – Probabilistic inference – Scientific inference – Other forms of inference n Integrating inference with – Data mining – Experimental processes

21. Phase Transitions and Undecidability