1 of 51

Ontologies 101: Standardize your data using ontologies�Nicole Vasilevsky�EHSRC seminar�September 10, 2021

2 of 51

Hello!

I am Nicole Vasilevsky

You can find me at:

nicole@tislab.org

@n_vasilevsky

2

3 of 51

Resources

Slides:

bit.ly/ehsrcseminar

Ontology Resources:

http://obofoundry.org/resources

3

4 of 51

1.

Introduction to Ontologies

4

5 of 51

Ontologies are systematic representations of knowledge that can be used to integrate and analyze large amounts of heterogeneous data

Defining entities and the relationships between them in a way that allows computational logical reasoning

5

DOI: 10.1056/NEJMra1615014

Image credit: https://kyndi.com/blog/creating-knowledge-and-maximizing-the-value-of-data-with-ontologies/

What

How

6 of 51

6

Terminology

A collection of terms, may include synonyms, definitions.

​

Ontology

A classification of knowledge in which terms are textually and logically defined. Provides semantics or meaning to terms.

​

Terminology vs Ontology

7 of 51

Complexity of Vocabulary Types

8 of 51

What can you do with a terminology?

Indexing with MeSH

8

Organizing data

Filtering data

Example:

9 of 51

What can you do with an ontology?

9

Organizing data

Filtering data

Connecting data

Suggesting data

10 of 51

Ontology use case

Website says sells light beer

Website says sells pilsner, lager, pale ale, etc.

Query

Where can I buy light beer?

Store #1

Store #2

11 of 51

Elements of an Ontology

11

Beer

is_a

is_a

is_a

is_a

is_a

is_a

Porter

Stout

Pilsner

Lager

Dark beer

Light beer

Classes (terms) - concepts in an ontology

Relationships (properties) - how the classes relate to each other

Instances - individuals

A knowledge classification of a domain, where the relationships between concepts are formally defined and logically related, which allows for computational reasoning

12 of 51

Ontology use case

Query

I want to find symptoms when someone was exposed to fertilizer.

13 of 51

Ontology use case

Query

I want to find symptoms when someone was exposed to fertilizer.

14 of 51

Examples

14

Domain

Usages

Open?

Biological processes, Cellular components, Molecular functions

Cross-species diseases, integrates many disease ontologies

Scientific investigations, e.g. assays, devices, objectives, and more

Abnormal phenotypes encountered in human diseases

Environmental exposures

Clinical terminology; diseases, symptoms, findings

Yes

Yes

Yes

Yes

Yes

Yes

No

Experimental variables including disease, chemicals, assays

Disease annotations in Monarch Initiative, ClinGen, Kids First, Ancestory.com

Disease annotations in Monarch Initiative, ClinGen, Kids First, Ancestory.com

eagle-i Network, Human Brain Project, Immune Epitope Database and Analysis Project

Genomic Standards Consortium, ESIP Federation, UN Environment, and IOC-UNESCO

Monarch Initiative, Electronic Health Records, The 100000 Genomes Project

EBI databases, including Open Targets

Electronic Health Records (EHR)

15 of 51

2.

Ontology Structure

15

16 of 51

Ontologies are expressed in formal language

like Web Ontology Language (OWL)

OWL is a semantic web computational logic-based language, designed to represent rich and complex knowledge about things and the relations between them. It also provides detailed, consistent and meaningful distinctions between classes, properties and relationships.

​

“Why not be inconsistent in at least one aspect of a language which is all about consistency?”

- Guus Schreiber, Why OWL and not WOL?

​

16

17 of 51

Key Features

  • Hierarchical classification: Terms are arranged in a hierarchy
  • Text definitions: Terms are defined
  • Synonyms: alternative terms
  • Logical definitions: Semantics - relationships between terms are defined, allowing logical inference and sophisticated data queries

HP:0002354

Memory impairment

​

Text def:

An impairment of memory as manifested by a reduced ability to remember things such as dates and names, and increased forgetfulness.

​

Synonyms:

Forgetfulness, Memory loss, Poor memory…

​

Logical def:

'has part' some

('increased rate'

and ('characteristic of' some forgetfulness)

and ('has modifier' some abnormal))

​

18 of 51

The is_a relationship

  • Terms (classes) in an ontology are often classified via an ‘is_a relationship’
  • The is_a relation is like inheritance
  • Children terms inherit the properties and relationships of the parent term

​

18

Dietary chemical entity exposure

An exposure event arising as a result of ingestion in food of chemical entity.

​

Dietary arsenic exposure

An exposure event arising as a result of ingestion in food of arsenic.

19 of 51

The is_a relationship

​

  • What is a subsumption hierarchy?
  • Which class is a superclass of ‘dietary food additive exposure?
  • Which class is a subclass of ‘food consumption related exposure’?
  • Is ‘dietary flavouring agent exposure’ a type of ‘dietary chemical entity exposure’?
  • What is the relationship of ‘dietary arsenic exposure’ to ‘dietary chemical entity exposure’?

​

19

20 of 51

Ontologies textually and logically define the relationships between classes

​

​

20

Logical (machine readable) definition

'exposure event'

and ('has exposure stimulus' some

('chemical entity'

and ('has role' some 'flavouring agent')))

and ('causally upstream of or within' some ‘Food Consumption')

Text (human readable) definition

An exposure event arising as a result of ingestion of food containing chemicals with the role of flavouring agent.

21 of 51

3.

Use cases: Ontologies usage in everyday life and disease diagnostics

21

22 of 51

Ontologies in Everyday Life

22

23 of 51

23

What is the Monarch Knowledge Graph?

  • monarchinitiative.org
  • Integrator of cross species genotype-phenotype data
  • OWL and DOS-DP
  • Uses OBO Foundry ontologies
  • obofoundry.org

doi.org/10.1093/nar/gkz997

24 of 51

24

Ontologies are used in disease diagnostics

  • Integrating data across species
  • Fuzzy matching
  • Rare disease diagnostics
  • Make inferences about humans

doi.org/10.1093/nar/gkz997

25 of 51

Human Phenotype Ontology (HPO)

​

​

https://hpo.jax.org/app/

​

​

Over 15,000 phenotype terms

​

​

26 of 51

26

Phenotype to disease annotations can be used for rare disease diagnoses

Disease to phenotype annotations

155,624 rare disease - phenotype annotations

136,268 common disease -phenotype annotations

27 of 51

Legend

Perfect Match

Fuzzy Match

No Match

  • Two patients with mutation in KMT2A
  • Not same variant
  • Phenotype similarity metric helped inform diagnosis of same disease

Differential diagnosis with similar but non-matching phenotypes is difficult

​

​

28 of 51

3.

How to look up ontology terms/hierarchies

28

29 of 51

29

https://www.ebi.ac.uk/ols/index

30 of 51

30

31 of 51

31

A library of ontologies and ontology related services

http://bioportal.bioontology.org/

32 of 51

32

Query ontologies, and provides RDF supporting remote query of each ontology term and the Semantic Web

http://www.ontobee.org

33 of 51

33

A community of ontologists committed to a shared set of principles to build open biomedical ontologies.

http://obofoundry.org/

34 of 51

34

  • There are multiple ontologies that exist, start by selecting the appropriate ontology, then search and restrict your search to that ontology.
  • Recommend using ontologies that are open and interoperable. Focusing on OBO foundry ontologies are a good place to start
  • Make informed decision about which ontology to use
  • Maybe the ontology you want to use does not have the term you want, so make a term request to that ontology

​

Which is the right ontology to use?

35 of 51

3.

Managing Ontologies in GitHub

35

36 of 51

36

  • Free, open-source online platform for hosting for software development projects
  • Provides version control using Git
  • Widely used in the ontology development community to:
          • store ontology files
          • version control
          • issue tracker

​

Sign up for free at: www.github.com

37 of 51

Ontology issue trackers

37

1

2

3

4

A tracker is a place to put a formal ontology request

​

Trackers have long been used in the software community for keeping track of bugs, feature requests, etc

Advantages:

  • Open
  • Documentation
  • Community can comment

​

Tracker IDs can be referenced in ontology metadata, such as in an editor note or definition annotation

​

38 of 51

38

https://github.com/EnvironmentOntology/environmental-exposure-ontology

39 of 51

ECTO GitHub Issue Tracker

39

https://github.com/EnvironmentOntology/environmental-exposure-ontology/issues

40 of 51

How do you write a tracker request?

  • Provide as much information as possible, in order to facilitate the change you are requesting and for future reference
  • For new terms, or term rearrangements, provide the intended hierarchy – both parent/subclass as well as any other relations required
  • Provide text definitions for all new or edited terms
  • Provide attribution for the definitions
  • Provide suggested synonyms

40

41 of 51

4.

Participate in Ontology Communities

41

42 of 51

Expert contributions to ontologies

43 of 51

Term search �and request workflow

Adapted from:

https://douroucouli.wordpress.com/2021/07/03/how-select-and-request-terms-from-ontologies/

Search using search strings

Return term list

Assume no results…

Examine list

Confirm, these are analogous to what you need

Check the parent ontology

Examine tickets

This is the term you are looking for

Check definitions and parent ontologies

Reads ticket

Make new term locally

Search for similar terms

Sibling terms list

Search for relevant discussion

Return tickets

Post term request

You

Portal

GitHub

Curator

You

Portal

GitHub

Curator

New term ID

Check

Term ID

Acknowledge

Make new ontology release

44 of 51

Summary

  • Add value to your data by annotating your data with external ontologies
  • Semantic enrichment aids in computational analysis
  • Participate in ontology communities, your contributions are appreciated and valued
  • Join OBO Foundry Slack: obo-communitygroup.slack.com

45 of 51

Acknowledgements

45

Melissa Haendel

University of Colorado

Chris Mungall

Lawrence Berkeley National Lab

David Osumi-Sutherland�European Bioinformatics Institute

Nico Matentzoglu

Semanticly

Some content was adapted from:

BDK14: Introduction to OWL2 and data reasoning | https://github.com/OHSUBD2K/BDK14-Ontologies-101

Anne Thessen

University of Colorado

46 of 51

Thanks!

You can find me at:

nicole@tislab.org

@n_vasilevsky

46

47 of 51

5.

Tools for Building Ontologies

47

48 of 51

Protege

48

Many ontologies are built using the software tool, Protégé

​

Can also use Protégé for viewing ontologies

https://protege.stanford.edu/

49 of 51

ROBOT

49

http://robot.obolibrary.org/

50 of 51

Ontology Development Kit (ODK)

50

https://github.com/INCATools/ontology-development-kit

PURPOSES

​

A toolkit to initialize an Open Biomedical Ontology (OBO) Library GitHub repository and associated files

​

​

    • GitHub repo with
      • Directory layout
      • Makefile
      • Imports
      • .travis.yml
      • README, License, Contributing, etc

​

GENERATES

51 of 51

Want to learn more about ontologies?

Online Ontology Tutorials

What: Basics of ontology development using software tool Protégé

Website: https://github.com/OHSUBD2K/BDK14-Ontologies-101

​

What: ROBOT tutorial

Website: http://robot.obolibrary.org/

​