Ontologies for biological annotation

1
(No Transcript)
2
GO terms implicitly refer to other term

• cysteine biosynthesis
• myoblast fusion
• hydrogen ion transporter activity
• snoRNA catabolism
• wing disc pattern formation
• epidermal cell differentiation
• regulation of flower development
• interleukin-18 receptor complex
• B-cell differentiation
• dorsal ectoderm

3
(No Transcript)
4
biosynthesis is_a metabolism
5
cysteine is_a serine family amino acid is_a amino
acid is_a amine
6
cysteine is_a serine family amino acid is_a amino
acid is_a serine
7
Composed terms currently cause problems

• No link to external ontology term
• Redundancy
• Inconsistency
• Extra work
• Annotation bottleneck
• Tangled DAGs and confusing displays
• we have no way to disentangle
• Solution so far
• fix errors based on results of term name parsing
  (Obol)
• reactive, not proactive

8
Solution actively manage composed terms

• Explicit pre-coordination
• Composed terms should now/soon be coordinated
  using oboedit plugin
• building block terms are recorded in ontology
  along with composite term
• Benefits
• Correct DAG structure can be inferred from
  external ontologies
• e.g. make sure GO CHEBI align
• placement consistency checking automated
• additional work can be automated
• synonyms, text definitions

9
How will terms be pre-coordinated by oboedit?

• How do we record a definition for a composite
  term?
• using a logical definition (computational
  essence)
• A logical definition consists of
• a generic term (aka genus)
• relationships to other terms which serve to
  discriminate this specific term from other is_a
  children of the generic term (aka differentiae)
• Can be written in natural language as
• A ltgeneric termgt which ltdiscriminating
  characteristicsgt

10
Example of pre-coordination

• cysteine biosynthesis
• generic term
• biosynthesis
• discriminating characteristics
• outputs cysteine
• natural language (Aristotelian style)
• a biosynthesis process which outputs cysteine

11
Example in Obo format
Term id GO0019344 name cysteine
biosynthesis intersection_of GO0009058 !
biosynthesis intersection_of outputs CHEBI15356
! cysteine is_a GO0009070 ! serine family amino
acid biosynthesis is_a GO0006534 ! cysteine
metabolism
12
Alternate syntax
GOcysteine_biosynthesis GObiosynthesis ?
outputs(CHEBIcysteine)

• used in pheno-syntax
• more compact
• similar to OWL abstract syntax
• I use Obo1.2 format or natural language in the
  rest of this presentation

13
This allows us to dynamically untangle

• Process axis view (primary is_as, via generic
  term)
• biological_process
• metabolism
• biosynthesis
• cysteine biosynthesis
• Process participant axis view
• amine
• amino acid
• serine family amino acid
• cysteine
• Combined view
• (same as current tangled diamond lattice)

14
Obol demo

• http//yuri.lbl.gov/amigo/obol

15
Recording the relationship is important

• Why not just a simple cross-product?
• e.g. biosynthesis x cysteine
• Relationships are important for reasoning and
  querying
• Consider
• cysteine biosynthesis from serine
• mRNA export from nucleus during heat stress
• Without the relations, the logical definition is
  not specific enough
• the essence is not captured
• Relations should come from RO
• more required

16
Multiple discriminating characteristics are
allowed

• Cysteine biosynthesis from serine
• Generic term
• biosynthesis
• Discriminating characteristics
• output cysteine
• input serine

Term name cysteine biosynthesis from
serine intersection_of GO0009058 !
biosynthesis intersection_of outputs CHEBI15356
! cysteine intersection_of input CHEBI17822 !
serine
17
Composite terms can be nested
Term id GOxxxxxxx name regulation of
cysteine biosynthesis intersection_of GO0050789
! regulation of biological process intersection_of
regulates GO0019344 ! cysteine
biosynthesis Term id GO0019344 name
cysteine biosynthesis intersection_of GO0009058
! biosynthesis intersection_of outputs
CHEBI15356 ! cysteine

• regulationregulates(biosynthesisoutputs(cyste
  ine))
• regulationregulates(biosynthesis)outputs(cystei
  ne)

YES
NO
18
Composite terms can optionally be manufactured in
bulk

• Generic term metabolism,biosynthesis
• Differentia has_output serine, cysteine,
• With caution
• Sparse vs dense matrices
• not all combinations are types

19
On the importance of necessary and sufficient
conditions

• Why intersection_of?
• Why not just make normal links in the GO DAG?
• normal relationships are for necessary conditions
  only
• we want both necessary and sufficient conditions
• captures the essence of the term

20
Normal DAG links only capture necessary
conditions, not essence
immune cell activation
inflammatory response
is_a
text def
macrophage activation
part_of
A change in morphology and behavior of a
macrophage resulting from exposure to a cytokine,
chemokine, cellular ligand, pathogen, or soluble
factor
21
Indistinguishable by DAG
immune cell activation
inflammatory response
is_a
text def
monocyte activation
part_of
A change in morphology and behavior of a monocyte
resulting from exposure to a cytokine, chemokine,
cellular ligand, pathogen, or soluble factor
22
essence captured by genus-differentia
immune cell activation
inflammatory response
is_a
macrophage activation
part_of
id GOmacrophage_activation intersection_of
GOcell_activation intersection_of activates
CLmacrophage
23
essence captured by genus-differentia
cell activation
is_a
immune cell activation
inflammatory response
genus
is_a
part_of
CLmacrophage
macrophage activation
activates
id GOmacrophage_activation intersection_of
GOcell_activation intersection_of activates
CLmacrophage
24
Current status of pre-coordinated terms

• SO already contains composite terms
• 46 pre-coordinated terms
• A silenced gene is a gene which has the quality
  of being silenced
• GO-BP/CL integration underway
• retrospectively pre-coordinated terms
• Obol page has pre-coordinated terms from
  automatic parsing
• http//www.fruitfly.org/cjm/obol

25
Pre- vs post- coordinated

• Pre-coordination
• terms are in ontology with IDs and computable
  definitions
• increases complexity of ontology
• complexity can be managed by tools
• e.g. new oboedit features
• Post-coordination
• terms are combined in the database
• forces more complexity in database schema and
  database applications

26
Pre-coordination is useful in moderation

• Commonly used terms should be pre-coordinated
• eg cysteine biosynthesis oocyte differentiation
  pectoral fin
• Avoid taking to extremes
• cf ICD-9
• Where do we draw the line?
• ontologies should be built around one or a few
  axes of classification
• term explosion typically gets large when
  multiple axes are combined
• we can change our minds later
• pre- and post- coordination is commensurable

27
Commensurability

• Annotator annotates to
• nucleuspart_of(astrocyte)
• Anatomy editor creates new term
• uses oboedit cross-product plugin
• astrocyte_nucleus nucleuspart_of(astrocyte)
• Annotation can be dynamically promoted to new
  term in answer to queries
• various software techniques for achieving this

28
Post-coordination in GO annotations

• Pre- and post- coordination are compatible and
  commensurable
• We should extend the annotation format to allow
  denoting more specific classes
• e.g.
• cholesterol transport in liver
• advanced applications can query this
• standard applications suffer no loss
• extended annotations can be used to help seed new
  terms in the ontology
• This is already being done (MGI,Dicty)
• we just want to capture this in interopeable way

29
Post-composition in gene association files

• New column in GA file format

30
Database issues

• Chado and GO DB can handle pre- and post-
  coordination
• in theory anyway
• not yet fully tested
• How does it work?
• anonymous term created for coordinated term
• documentation in chado cvs
• chado/modules/cv/doc/