Towards a Computational Paradigm for Biological Structure

1
Towards a Computational Paradigm for Biological
Structure
First International Workshop on Formal Biomedical
Knowledge Representation (KR-MED2004), June 1,
2004, Whistler (Canada)

• Stefan Schulz
• Department of Medical InformaticsUniversity
  Hospital Freiburg (Germany)

Udo Hahn Text Knowledge Engineering
LabUniversity of Jena (Germany)
2
The World of Life Sciences
Millions of Species
Evolutionof Life
Function
Morphology
Organs
Dysfunction
Organ Systems
Organisms
Molecules
Genes
Tissues
Cells
3
requires sophisticated organization
Bio-ontologies
4
What exists ?

• Human Anatomy
• Foundational Model of Anatomy (FMA)
• Portions of SNOMED, OpenGalen, MeSH
• Other Organisms
• Open Biological Ontologies (OBO)
• Mouse (developmental stages), Zebrafish,
  Drosophila,
• Species-Independent
• Gene Ontology Cellular Component

5
Overlap
6
Same name different meaning

• Motor Neuron instance-of Neuron (FlyBase)

• Motor Neuron narrower Neuron (MeSH)

• Motor Neuron subclass-of Neuron (FMA,
  OpenGALEN)

7
Same name different meaning

• Cell has-part Axon (Gene Ontology)
• Do cells without axons exist ?
• Do axons withoutcells exist ?

• Neuron has-part Axon (FMA)
• Does every neuron has an axon?

8
Deficiencies (II)
Keep in mind that part_of means can be a part
of, not is always a part of GO Editorial Style
Guide, Oct 2003 The part_of relationship () is
usually necessarily is_part GO Editorial Style
Guide, Jan 2004

• Cell has-part Axon (Gene Ontology)
• Do cells without axons exist ?
• Do axons withoutcell exist ?

• Neuron has-part Axon (FMA)
• Does every neuron has an axon?

A part_of B if and only if for any instance x
of A there is some instance y of B which is such
that x stands to y in the instance-level part
relation, and vice versa. Rosse Smith MEDINFO
2004
9
Conflicting and / or underspecified
conceptualizations hamper sharing and
integration of ontologies
10
Semantic framework for biological structure

• Foundational Relations
• General Attributes
• Theories

11
Semantic framework for biological structure

• Foundational Relations
• General Attributes
• Theories

12
Foundational Relations between Biological
Structure

• is-a
• instance-of
• part-of / has-part
• has-location / location-of
• has-branch / branch-of
• has-developmental-form /is-developmental-form-of
• descends-from /has-descendant
• connects
• bounds / bounded by

classify bydomain / range ?
13
Two kinds of entities

Domain Entities

• Hand
• Blood
• Cell

• my left hand
• a blood sample
• a concrete cell

Universals(Concepts, Classes ofIndividuals)
Individuals (Concrete Objects)
14
Classification of Foundational Relations
part-of has-location has-branch has-developmental-
form bounds connects

Individuals (concrete objects)
15
Classification of Foundational Relations

Individuals (concrete objects)
16
From Instance-to-Instance relations to
Class-to-Class Relations

• A, B are classes, inst-of class membership
• rel relation between instances Rel relation
  between classes
• Rel (A, B) def
• ?x inst-of (x, A) ? inst-of (y, B) ? rel (x,
  y) OR
  ? x inst-of(x, A) ? ? y inst-of (y, B) ? rel
  (x, y) OR
• ? y inst-of(y, B) ? ? x inst-of (x, A) ? rel
  (x, y)

?
?
?
cf.Schulz Hahn (KR 2004, June 2, 11am)Rosse
Smith (MEDINFO 2004)
17
Semantic framework for biological structure

• Foundational Relations
• General Attributes
• Theories

18
General Attributes (mutually disjoint classes)

• Dimensionality Point, 1-D, 2-D, 3-D

19
Semantic framework for biological structure

• Foundational Relations
• General Attributes
• Theories

20
Theories

• A set of formal axioms which describe a
  restricted (local) domain.
• Four orthogonal theories for Biological Structure

• Granularity
• Species
• Development
• Canonicity

21
Theories

• A set of formal axioms which describe a
  restricted (local) domain.
• Four orthogonal theories for Biological Structure

• Granularity
• Species
• Development
• Canonicity

22
Granularity

• Level of detail (molecular, cellular, tissue,
  organ)
• Change in Granularity level may be non-monotonous
• Change of sortal restrictions
• 3-D ? 2-D boundary
• Count concept ? Mass concept
• Change of relational attributions
• disconnected ? connected

23
Theories

• A set of formal axioms which describe a
  restricted (local) domain.
• Four orthogonal theories for Biological Structure

• Granularity
• Species
• Development
• Canonicity

24
Linnean Taxonomy of Species
http//tolweb.org
25
Linnean Taxonomy of Species
http//tolweb.org
26
Linnean Taxonomy of Species
http//tolweb.org
27
Species

• Introduction of Axioms at the highest common level

Has-Part Skull
Has-Part Skull Has-Part Vertebra
Has-Part Skull Has-Part Vertebra Has-Part Jaw
28
Theories

• A set of formal axioms which describe a
  restricted (local) domain.
• Four orthogonal theories for Biological Structure

• Granularity
• Species
• Development
• Canonicity

29
Development

• Represents time-dependent snapshots from the
  life cycle of an organism, e.g.,zygote, embryo,
  fetus, child, adult
• Granularity stages are species-dependente.g.
  metamorphosis

30
Theories

• A set of formal axioms which describe a
  restricted (local) domain.
• Four orthogonal theories for Biological Structure

• Granularity
• Species
• Development
• Canonicity

31
Canonicity

• Degrees of Wellformedness of Biological
  Structure
• Canonic structure

32
Canonicity

• Degrees of Wellformedness of Biological
  Structure
• Canonic structure
• Structural Variations

33
Canonicity

• Degrees of Wellformedness of Biological
  Structure
• Canonic structure
• Structural Variations
• Pathological Structure

34
Canonicity

• Degrees of Wellformedness of Biological
  Structure
• Canonic structure
• Structural Variations
• Pathological Structure
• Lethal Structure

35
Canonicity

• Degrees of Wellformedness of Biological
  Structure
• Canonic structure
• Structural Variations
• Pathological Structure
• Lethal Structure
• Derivates of biologicalstructure

36
Canonicity

• Five canonicity levels each level introduces
  axioms valid for higher levels

37
Examples
Granularity
Species
Development
Canonicity
38
CoverageFoundational Model of Anatomy
Granularity
Species
Development
Canonicity
39
CoverageGene Ontology
Granularity
Species
Development
Canonicity
40
CoverageMouse Anatomy
Granularity
Species
Development
Canonicity
41
Examples

• Connects(RightVentricle, Left Ventricle)

Granularity normal Species
mammal Development adult Canonicity 4-5
false
Granularity any Species vertebrate Developmen
t early embryo Canonicity any
true
42
Conclusion

• Integration of bio-ontologies requires
• Uncontroversial semantics of relations and
  attributes
• Clear commitment to theories, such as
  granularity, species, development and canonicity
• Redundancy can be avoided
• Encoding axioms at the highest common level in
  the species taxonomy (e.g. vertebrates,
  arthropods, primates) and benefit from
  inheritance in subsumption hierarchies

43
(No Transcript)
44
requires sophisticated organization

• Formalization and Standardization of Clinical
  Terminologies
• Basis for the Annotation of Genes and Gene
  Products
• Semantic reference for scientific communication
• Machine-supported reasoning and decision-support

Bio-ontologies !
45
(No Transcript)
46
(No Transcript)
47
Upper level classification of entities
Individuals (concrete objects)
Universals(Concepts, Classes ofIndividuals)
Continuants(physical objects,)

• my left hand
• a blood sample
• a concrete cell

• Hand,
• Blood
• Cell

Occurrents (events, processes, actions)

• Peters diabetes
• appendectomy of Patient 12345

• Diabetes mellitus
• Appendectomy

48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
Mereotopological Quiz
??
??

• Glioblastoma has-location BrainGlioblastoma
  part-of Brain?

??

• Brain metastasis has-location Brain Brain
  metastasis part-of Brain?

??
??
Images from Sobotta CD-ROM
54
part-of has-location
transitive closure by taxonomic subsumption
55
Subtheories of an Ontology of Biological Structure

• 1. Taxonomy

2. Mereology
3. Topology
is-a
part-of
connection
Hand
part-of
Thumb
part-of
Thumbnail
56
Subtheories of an Ontology of Biological Structure

• 1. Taxonomy

2. Mereology
3. Topology
is-a
part-of
connection

• Canonical relationships

(Schulz et al. AMIA 2000)

• Topological Primitives

57
Structure of Talk

• Introduction
• Foundational Relations
• Foundational Attributes
• Theories
• Granularity
• Species
• Development
• Canonicity

58
The World of Life Sciences
59
Generalized Representation of Living Systems Top
Level
Biological Entities
Biological Occurrents process, state, event,
Biological Continuants organism, organ, tissue,
cell, molecule,..
dependence
60
Ontological Account for Biological Continuants

• Foundational Relations
• Foundational Attributes
• Theories
• Granularity
• Species
• Development
• Canonicity

61
Granularity

• Taxonomic degreeof specialization

molecular level, cellular level, tissue level,
organ level, population level
62
Change in Granularity level may be non-monotonous

• Change of sortal restrictions
• 3-D ? 2-D boundary
• Count concept ? Mass concept
• Change of relational attributions
• disconnected ? connected