Biological Ontologies

Neocles LeontisApril 20, 2005

What Is An Ontology?

• An ontology is an explicit description of a domain of knowledge:– Concepts -- Entities and Relations– Properties and attributes of Entities and Relations– Constraints on properties and attributes– Individuals (“Instances”)

• An ontology defines: – a common vocabulary– a shared understanding of the domain of knowledge– Commitments on how to use the vocabulary

What Is “Ontology Engineering”?

Ontology Engineering: Defining terms in the domain and relations among them– Defining concepts in the domain (Classes)– Arranging the concepts in a hierarchy

(Subclass-Superclass hierarchy)– Defining which attributes and Properties

classes can have (slots) and constraints on their values (facets)

– Defining individuals and filling in slot values (instantiation)

Why Develop an Ontology?

• To share common understanding of the structure of information – among people– among software agents

• To enable reuse of domain knowledge– to avoid “re-inventing the wheel”– to introduce standards to allow

interoperability between ontologies

More Reasons…• To make domain assumptions

explicit– easier to change domain assumptions– easier to understand and update

legacy data• To separate domain knowledge

from the operational knowledge– re-use domain and operational

knowledge separately

An Ontology Is Often Just the Beginning

OntologiesOntologies

Problem-solving

methods

Problem-solving

methods

DatabasesDatabasesDeclarestructure

Knowledgebases

Knowledgebases

Providedomain

description

Ontology-Development Process

In Logical order:

determinescope

considerreuse

enumerateterms

defineclasses

defineproperties

defineconstraints

createinstances

In reality - an iterative process:

determinescope

considerreuse

enumerateterms

defineclasses

considerreuse

enumerateterms

defineclasses

defineproperties

createinstances

defineclasses

defineproperties

defineconstraints

createinstances

defineclasses

considerreuse

defineproperties

defineconstraints

createinstances

Protégé

• Graphical ontology-development tool

• Supports a rich knowledge mode• Open-source and freely available

(http://protege.stanford.edu)

Authoring Program (Protégé 2000)

• Enforces the implementation of foundational principles and definitional desiderata

• Frame-based architecture compatible with OKBC protocol = Open Knowledge Base Connectivity

• Frames are used to represent anatomical concepts

• Frames allow for distinguishing between class and instance

• Protégé allows for selective inheritance of attributes

• Protégé enhances specificity and expressivity of attributes by assigning them their own attributes.

Determine Domain and Scope

• What is the domain that the ontology will cover?

• Who is going to use the ontology?• For what are they (we) going to use the

ontology?• For what types of questions should the

information in the ontology provide answers (competency questions)?

Answers to these questions may change during the lifecycle

determinescope

considerreuse

enumerateterms

defineclasses

defineproperties

defineconstraints

createinstances

RNA Ontology Scope: DOMAIN

– RNA Sequences (1D) -- Coding and Non-Coding

– RNA 2D structures – RNA 3D structures– Alignments of homologous RNA sequences– Relationships between alignments and 3D

structures

RNA Ontology ScopeWHO?

– Molecular biologists & biochemists – Structural biologists – Evolutionary biologists– Nanotechnologists

RNA Ontology Scope: WHAT?

– How to improve prediction of RNA 3D structure

– How to improve sequence alignments of homologous RNAs

– To identify and annotate RNA genes in genomes

– How are RNA 3D structure and evolution coupled?

– How is RNA evolution coupled to biological evolution

Consider Reuse

• Why reuse other ontologies?– to save the effort– to interact with the tools that use

other ontologies– to use ontologies that have been

validated through use in applications

determinescope

considerreuse

enumerateterms

defineclasses

defineproperties

defineconstraints

createinstances

Enumerate Important Terms

• What are the terms (entities) we need to talk about?

• What are the properties and attributes of these entities?

• What are the relationships between entities?

considerreuse

determinescope

enumerateterms

defineclasses

defineproperties

defineconstraints

createinstances

Define Classes and the Class Hierarchy

• A class is a concept in the domain– a class of wines– a class of wineries– a class of red wines

• A class is a collection of elements with similar properties

• Instances of classes– a glass of California wine you’ll have for lunch

considerreuse

determinescope

defineclasses

defineproperties

defineconstraints

createinstances

enumerateterms

Class Hierarchy

Class Inheritance

• Classes usually constitute a taxonomic hierarchy (a subclass-superclass hierarchy)

• A class hierarchy is usually an IS-A hierarchy:

an instance of a subclass is an instance of a superclass

• If you think of a class as a set of elements, a subclass is a subset

FMA -- High Level Scheme

• FMA = (AT, ASA, ATA, Mk)– AT = Anatomy taxonomy (assigns anatomical

entities as class concepts– ASA = Anatomy Structural Abstraction --

includes structural relationships among entities of the AT

– ATA = Anatomical Transformation Abstraction -- relationships that describe morphological & physical transformations of anatomical entities

– MK = Metaknowledge -- principles and sets of rules

ASA -- High Level Scheme

• ASA = (Dt, PPt, Bn, Pn, SAn)– Dt = Dimensional taxonomy– PPt = Physical Properties taxonomy– Bn = Boundary network– Pn = Partonomy network– SAn = Spatial Association network

Boundary Network (Bn)

• Specification of boundaries is critical for segmentation of images and volumetric datasets

• Definition: Boundary = Non-material physical anatomical entity of two or fewer dimensions that delimits anatomical entities that are of one higher dimension than the bounding entity

Boundary Network (Bn)

Inverse Relationships: -bounded by--bounds-

Real vs. Virtual Boundaries:Rea boundaries correspond to its surface and designate discontinuities between constitutional parts of anatomical entities

Partonomy Network (Pn)

Inverse Relationships: -has part-

Rule of Dimensional Consistency

Distinguishes between boundary and partonomy relationships.

Parthood relations -- only allowed for entities of the same dimension

Ex: Cavity of stomach (3D) -has part- Cavity of pyloric antrum (3D)

Ex: Internal surface of stomach (2D) -has part- Internal surface of pyloric antrum (2D)

What to Reuse?

• Ontology libraries– DAML ontology library

(www.daml.org/ontologies)

– Ontolingua ontology library (www.ksl.stanford.edu/software/ontolingua/)

– Protégé ontology library (protege.stanford.edu/plugins.html)

• Upper ontologies– IEEE Standard Upper Ontology

(suo.ieee.org)– Cyc (www.cyc.com)

RNA Ontology Consortium

• To share common understanding of the structure of information – among people– among software agents

• To enable reuse of domain knowledge– to avoid “re-inventing the wheel”– to introduce standards to allow

interoperability

What to Reuse? (II)

• General ontologies– DMOZ (www.dmoz.org)

– WordNet (www.cogsci.princeton.edu/~wn/)

• Domain-specific ontologies– UMLS Semantic Net– GO (Gene Ontology) (www.geneontology.org

)– FMA (Foundational Model of Anatomy)

Foundational Model of Anatomy

QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.

http://sig.biostr.washington.edu/projects/fm/index.html

• Reference ontology for biomedical informatics

• Representation of Anatomical Entities and Relationships

• Symbolic modeling of the structure of the human body at the highest level of granularity

• Evolving Resource for knowledge-based applications requiring anatomical information

FMA: Modeling Challenges• Representing complex structural relations• Representing different levels of granularity• Developing a model that is scalable to a

very large number of concepts• Using consistent organizational principles