Transcript Disease classification workshop Georgetown April 2012

Defining Disease Across Organisms
Buffalo PRO-PO-GO
May 2013
Judith Blake
Jackson Laboratory
Why do we need a formal disease
classifications?
To search tagged data
To aggregate datasets
To mine data sources; eg literature, EHRs
To search coded data for sub- and superclasses
To discover novel relationships between diseases within a species
To discover the relationships between diseases and pathways
To search for related diseases between species
To allow stratification of disease populations
Modified from Paul Schofield 2012 ISB presentation
Classification systems
1. those that belong to the Emperor
2. embalmed ones
3. those that are trained
4. suckling pigs
5. mermaids
6. fabulous ones
7. stray dogs
8. those included in the present
classification
9. those that tremble as if they were mad
10. innumerable ones
11. those drawn with a very fine
camelhair brush
12. others
13. those that have just broken a flower
vase
14. those that from a long way off look
like flies.
The Celestial Emporium of Benevolent Knowledge Jorge-Luis Borges
“Defining Disease in the Genomics Era”
• A disease is:
• a state that places individuals at increased risk of
adverse consequences.
• Where is the threshold for ‘adverse’ consequences
(a) along the intensity dimension; (b) along the time
dimension?
Science 293 (5531) I3 Aug 2001, pp. 807-808.
Modified from Barry Smith 2011 DO meeting
Disease
Cell
Anatomy
Adapted from Schriml and Kibbe: ICBO submission 2013
What are axis and intersections for crossorganism disease representations?
• Diseases and Phenotypes
– Phenoscape, Monarch, MGI, PhenomeNet
• Diseases and Taxonomic Distribution
– Taxonomies, Anatomies (Uberon)
• Diseases and Genotypes
– Inheritances, Complex Genotypes, Penetrance and
Susceptibility
• Diseases and Exposures
– Chemicals, Pathogens, ENV, EXO, CTD
Epidemiology Metadata
Define and Standardize:
Pathogen, Host, Reservoir, Mode of Transmission, Portal of Entry, Vector,
Disease, Symptom, Geographic Location
Lynn Schriml 2013
Diseases and Phenotypes
•
Diseases are (traditionally) described by signs and symptoms
– Signs – things you can measure
– Symptoms – things the patient notices
•
Signs are phenotypes
•
Diseases are characterized by phenotypes, the order, severity and duration with
which they occur. A full model of disease takes into account dimensions of
anatomy, time, severity, therapeutic responsiveness, outcomes etc. There is also a
probabilistic element to an instance of the disease and a probabilistic association
between phenotypic elements in one instance.
•
Diseases are not phenotypes ( although predisposition may be considered as
such) but single phenotype diseases may be viewed as phenotypes, eg.
Osteoarthritis or plant rust diseases.
Human Phenotype Ontology
• ≈ 10,500 terms
• ≈ 60,000 annotations for
mainly Mendelian disease
Broad uptake in human genetics community
http://www.human-phenotype-ontology.org
Slide courtesy of
Peter Robinson
Comparative Disease Phenotypes
• systemic disorder of connective tissue
• aortic aneurysm
• partial loss of microfibrillar function








(top) wild-type littermates
(bottom) Fbn1tm2Rmz/ Fbn1tm2Rmz
dolichostenomelia
arachnodactyly
micrognathia
abnormal chest/rib overgrowth
aortic aneurysm
decreased muscle mass
kyphosis
premature death
Marfan Syndrome:
caused by mutations in fibrillin 1 gene
Formalisation and logical definitions
• PATO, EQ syntax
• Phenomenet
• Mousefinder
Querying across species and time
UBERON
CL
NCIt
GO
EMAP
EDHAA2
ZFA
MA
FMA
SNOMED
many other
organism
ontologies
NESCent
(Vision, Lapp,
Balhoff, Kothari)
Working groups
Curator interface
U. Oregon
(Westerfield)
Usability testing
Database
Liason to ZFIN
Public interface
Liason to NCBO
U. South Dakota
Acad. Natural Sciences
(Mabee,Lundberg, Dahdul)
Morphology
collaborators
(Arratia, Coburn,
Hilton, Mayden)
Ostariophysan
phenotypic
data
Zebrafish
phenotypic
& genetic
data
Ontologies
(taxonomy, TAO,
PATO, homology)
NCBO
Applications
(Phenote, OBO-Edit)
OBO
(host of TAO, PATO,
taxonomy ontology)
Kansas
(Midford)
Ichthyology community
(DeepFin, Fishbase)
Ontology Curation
Phenotype Ontologies
for Evolutionary Biology
Workshops
Todd Vision 2013
Taxon phenotype annotations
Links a quality to
the entity that is
its bearer
Brachyplatystoma
capapretum
Taxon
ontology term
exhibits some
Anatomy
ontology term
Phenotypic Quality
ontology term
round that
inheres_in some
ethmoid
cartilage
Todd Vision 2013
sequence-specific DNA
binding transcription
factor activity
chondrocranium
cartilage
olfactory
region
Pimelodidae
shape
has_function
is_a
is_a
ethmoid
cartilage
Brachyplatystoma
variant_of
inheres_in
is_a
exhibits some
is_a
is_a
tfap2a
Brachyplatystoma
capapretum
part_of
round that
inheres_in some
ethmoid cartilage
tfap2ats213/ts213
influences some
round
is_a
split
inheres_in
is_a
split that
inheres_in some
ethmoid cartilage
Todd Vision 2013
Defining diseases and phenotypes
•
Phenotypic definition of a disease permits sophisticated computational
analysis within a species. However, disease classification necessary for
community tagging and tracking.
•
Formal definition of phenotypes allows cross-species data integration and
analysis. Plant Anatomy Ontology essential to phenotype classifications
•
The accuracy of the asserted hierarchy is essential for utility.
•
Inaccurate structure, inappropriate relations or incomplete classes render a
formal ontology worse than useless.
•
For asserted structures the upper levels need to reflect the uses to which it
will be put. Terminologies should be familiar to pathologists, plant
biologists and other major users
IDO scales of granularity
• Host-pathogen interactions
occurs at a variety of scales
Surveillance
– Ecosystem
– Organism
Pathogenesis
– Organ
– Cell
– Molecule
Richard Scheuermann 2011
Infections Disease Ontology scales & branches
Scale
Independent
Continuant
Dependent Continuant
Occurrent
Ecosystem
specimen isolation source,
environment
vector (role), carrier (role), population
density, routes of migration (?), host
(role), pathogen (role), source (?),
temperature
transmission, migration,
specimen isolation process,
specimen isolation event
Organism
NCBI taxonomy, vaccine
preparation,
healthy, sick, animal model (role), viral
load, latent (state), sterile eradication
(state), symptoms (quality), immunogen
(role), adjuvant (role), pathogen
(disposition), resistant (disposition)
pathogenesis, reactivation,
progression, immune
response, vaccination
Organ
FMA (mucosal organ - lung,
secondary lymphoid organ lymph node), pericavitary
tissue, abssess
viral load (quality), inductive site (role),
effective site, inflamed (quality),
granuloma (quality), caseous necrosis
(quality)
inflammation, tissue
damage, necrosis
Cell
T cell, dendritic cell
Infected (quality), activated, susceptible
(disposition)
antigen presentation,
proliferation, phagocytosis,
cytoxicity, apoptosis
Molecule
Glyoxalase, catalase, recA
detoxification (role), DNA lesion
recognition, DNA repair, transport
catalysis, binding, transport
Richard Scheuermann 2011
Tier 5
Formal
disease model
Tier 4 Diseases
Defined using formal
logical definitions
Tier 3b Diseases
Defined by
Phenotype terms plus
quantitative
and deep phenotyping data
Discovery
Tier 3a Diseases defined with
by Phenotype terms from an ontology
Tier 2 Ontology/DAG
Indexing and simple
searching
Tier 1 Controlled Vocabulary/Terminology
Paul Schofiield 2013
Tier 5
Formal
disease model
Work
Tier 4 Diseases
Defined using formal
logical definitions
Tier 3b Diseases
Defined by
Phenotype terms plus
quantitative
and deep phenotyping data
Tier 3a Diseases defined with
by Phenotype terms from an ontology
Tier 2 Ontology/DAG
Tier 1 Controlled Vocabulary/Terminology
Paul Schofiield 2013
Start from here?
• Review existing classifications: investigate, possibly emulate –
–
–
–
–
–
IDO (formal disease classification)
MEDIC (disease classification built on OMIM base)
Phenoscape (phenotype integration across species, anatomy-centric)
PhenomeNet
Orphanet (genetic disease classification)
Plant disease classifications
• Need extensive involvement of pathologists and biologists as well as
informaticians
• Recognize differences and requirements of disease vs phenotype
components, support inter-ontology constructions
• Do we need to reinvent the wheel or just pump up the tires?