Transcript IDO - Buffalo Ontology Site

BFO-aligned Ontologies for Clinical
and Translational Research:
OGMS, IDO, and VO
(Orlando Presentation, 2/8/2013)
http://ncorwiki.buffalo.edu/index.php/CTSA_Ontology_Workshop
Yongqun “Oliver” He
University of Michigan Medical School
Ann Arbor, MI 48109
BFO and OBO Foundry Principles
• BFO: Basic Formal Ontology
• BFO has been used a top level ontology for many
ontologies associated with clinical and translational
research
• Examples: OGMS, IDO, VO, OBI, OAE
• All OBO foundry library ontologies follow OBO
Foundry principles, e.g., openness, collaboration, and
use of a common shared syntax
OGMS:
Ontology of General
Medical Science
• An ontology of the major
types of entities involved
in a clinical encounter.
– An upper ontology for
clinical medicine
– A mid-level ontology with
respect to BFO
• Includes ~100 general
terms
• By:
– Albert Goldfain
– Richard Scheuermann
– Barry Smith, …
https://
code.google.com
/p/ogms/
Wide OGMS Applications
Ontologies
using OGMS:
•
•
•
•
•
•
•
•
•
Courtesy: figure kindly provided by Albert Goldfain
IDO
DO
SDO
AERO
OAE
VSO
OMRSE
VO
...
OGMS application example:
Development of OAE
• OAE: Ontology of Adverse Events
• OAE ‘adverse event’:
– = def. a OGMS: ‘pathological bodily process’
that occurs after a medical intervention.
– Does not assume causality
– ‘causal adverse event’ assumes causality
• >1,000 specific AE terms in OAE now,
mapped to MedDRA terms
OAE for AE data analysis
Ref: Sarntivijai et al., 2012
PLoS ONE
IDO: Infectious Disease Ontology
IDO-core Central Terms:
• IDO:
represent the
entire
infectious
disease
domain
• Interoperability
with other
disease and
health domains
• IDO-core: by
Lindsay Cowell,
Barry Smith,
and others
Courtesy: figure kindly provided by Lindsay Cowell
IDO Core-Extension Development Strategy
• IDO extensions are developed by extending IDO-core
IDO-Asp
IDO-Sa
OGMS
CL
IDOCry
IDOFun
I
GO BP
IDOBac IDOTB
IDO-Core
IDOFlav IDOVirus
IDO-Flu
IDOPar
IDOSch
OBI
IDO-Mal
Courtesy: figure kindly provided by Lindsay Cowell
• We developed an IDO extension: Brucellosis Ontology
IDOBRU: Brucellosis Ontology as an
IDO Extension
• Focuses on the domain of zoonotic brucellosis, caused by Gramnegative bacterium Brucella.
• Incorporates all IDO-core terms, has over 880 Brucella-specific
terms, and imports terms from other ontologies.
Citation: “Asiyah” Yu Lin, Zuoshuang Xiang, Yongqun He.
Brucellosis Ontology (IDOBRU) as an extension of the Infectious Disease Ontology.
Journal of Biomedical Semantics. 2011 Oct 31;2(1):9. PMID: 22041276.
IDO-core is the top ontology of IDOBRU
Vaccine Ontology (VO)
• VO: A biomedical ontology in the
domain of vaccine and vaccination
• Support data integration, literature
mining, and reasoning
• Integrated with VIOLIN
• VIOLIN: a vaccine database and analysis
system, including many programs, e.g.:
o
o
o
o
o
~3000 vaccines
Protegen: protective antigens. ~600
Vaxjo: vaccine adjuvants: > 100
Vaxign: vaccine design
Widely used by vaccine community
http://www.violinet.org
• Funded by a NIAID R01 grant
http://www.violinet.org/vaccineontology
Many Ontology Tools developed
during VO development
Linked ontology data server
Ontology fetching tool
Ontobee:
linked ontology
data server
OntoFox: reuse
existing ontology
terms
is_a
ontology
development
tool
is_a
Ontorat: generate
new ontology
terms
is_a
Hegroup
RDF triple
store
uses
Ontodog: generate
ontology community
view
Mass
generation of
new terms
is_a
uses
uses
Ontology
community
view generator
uses
Ontobat: biodata
analysis tool
(In development)
is_a
is_a
uses
OntoCOG:
COG enrichment
analysis tool
Ontology data
analysis
ontology
application
tool
VO Statistics and Development
#
Class
VO
BFO 2
RO
CARO
CHEBI
DOID
GO
OBI
OGMS
PATO
FMA
IAO
IDO
NCBITaxon
PRO
UBERON
UO
4800
22
0
9
20
57
19
36
1
17
2
18
2
397
2
8
1
Object
Property
7
38
4
0
0
0
0
11
0
0
0
2
0
0
0
0
0
Subtotal
5411
62
Subtotal
4807
60
4
9
20
57
19
47
1
17
2
20
2
397
2
8
1
5473
• OntoFox to import
external terms and
axioms from other 16
ontologies
• Ontorat to generate a
large number of terms
and axioms automatically
• VO includes >1000
vaccines for >20 host
spp. against various
diseases
VO imports OBI terms for vaccine investigation
OBI/VO modeling of
“vaccine protection
assay”
OBI:
Ontology for Biomedical
Investigations
~20 communities involved
Reference: Brinkman et al. (2007). Modeling biomedical experimental processes with OBI.
Journal of Biomedical Semantics. 2010, 1(Suppl 1):S7. PMID: 20626927.
Example: Afluria Influenza Vaccine
age
has_
quality
bearer
_of
Bob (a
human)
Influenza virus
vaccine
host role
bearer_of
quality_is_measured_as
age measurement
datum (value: 6
unit: month)
has_
participant
plan
specification
has_part
dose
specification
viral pathogen
target role
intramuscular
vaccination
is_a
measurement
data
realizes
is_about
is_
realized
-by
has_
quality
inactivated
adaptive immune
response
has_participant
realizes
has_part
is_specified_
input_of
Afluria-1
has_participant
viral vaccineinduced
immunization
has_
bearer_of some
specified
_output ‘acquired immunity
to Influenza virus’
_of
is_a
Flu vaccine
has_part
is_
manufactured
_by
CSL Limited
chicken egg
protein allergen
bearer_of
vaccine
allergen
disposition
U of Michigan Ontology Research
•
UM Ontology Working Group:
o
o
o
•
UM MCubed pilot award:
o
o
•
Title: Ontology Development and Applications for Clinical
and Translational Science
To: Alla, Marcy, and Oliver; Period: 1.5 years
UM CTSA: Michigan Institute for Clinical & Health
Research (MICHR)
o
o
o
•
Members: Marcy Harris, Alla Karnovsky, Frank Manion,
Oliver He, Asiyah Yu Lin, Jeff Cowall, …
Activities: Biweekly meetings, …
Developing a Clinical and Translational Research Ontology.
Ontology research needed to integrate huge datasets
Committed to collaborative community effort
One project: Informed Consent Ontology (ICO) (next slide)
Case study: Head and neck cancer biorepository
Informed Consent Ontology (ICO)
•
•
•
ICO: A prototype, aligned with BFO.
Currently focused on research permissions
UM CTSA Project Team:
Alla Karnovsky, Frank Manion, Marcy Harris,
Oliver He, Nick Steneck, Blake Roessler
Patient
Record
Institutional
Records
IRB/
eResearch
Protocol
Informed
Consent Form
Courtesy:
figures
kindly
provided by
Alla
Karnovsky
and
Nickolas
Steneck
Subject matter expert view
Reference: Development of an Informed Consent Ontology to Support Biobanking. Alla Karnovsky,
Frank J. Manion, Oliver He, Terry Weymouth, V. Glenn Tarcea; Lisa Powell, Blake J. Roessler,
Nicholas H. Steneck. AMIA 2012 Annual Symposium.
Clinical Data Integration Required
•
•
Records of millions of patients in UM Health System (UMHS)
Ontology is needed for true clinical data integration
Courtesy: figure kindly provided by Jeff Cowall
Acknowledgements
Oliver He Group:
• Zuoshuang “Allen” Xiang
• “Asiyah” Yu Lin
• Sirarat Sarntivijai
UM Literature Mining
Collaborators:
• Arzucan Özgür
• Junguk Hur
VO Collaborators:
Barry, Lindsay, Alan, Bjoern, ….
UM Ontology Working Group
Listed in a previous slide
Barry Smith
(BFO, OGMS, IDO, VO, ...)
Lindsay Cowell (IDO)
OGMS Development Team
• Albert Goldfain
• Richard Scheuermann, …
OBI Consortium:
•
•
•
•
•
Bjoern Peters
Jie Zheng
Chris Stoeckert
Alan Rutternberg
Melanie Courtot, …
Funding: NIH grants R01AI081062 & U54-DA-021519
UM MCubed pilot project, MICHR (UM CTSA)