Difference between revisions of "Entities with taxonomic context"

From phenoscape
(Phenotype annotations)
(Phenotype annotations)
Line 2: Line 2:
  
 
==Phenotype annotations==
 
==Phenotype annotations==
 +
===Old form===
 
Our current phenotype annotation model has the form "a particular taxon has some phenotype, consisting of some quality inhering in some anatomical entity":
 
Our current phenotype annotation model has the form "a particular taxon has some phenotype, consisting of some quality inhering in some anatomical entity":
'''Old form'''
 
 
  Danio rerio ''exhibits'' sigmoid^''inheres_in''(vertebra 1)
 
  Danio rerio ''exhibits'' sigmoid^''inheres_in''(vertebra 1)
 
The phenotype is a post-composition with the following properties:
 
The phenotype is a post-composition with the following properties:
 
  sigmoid^''inheres_in''(vertebra 1) ''is_a'' sigmoid
 
  sigmoid^''inheres_in''(vertebra 1) ''is_a'' sigmoid
 
  sigmoid^''inheres_in''(vertebra 1) ''inheres_in'' vertebra 1
 
  sigmoid^''inheres_in''(vertebra 1) ''inheres_in'' vertebra 1
 +
===New form===
 
This annotation can be stated differently as "some anatomical entity, in a particular taxon, has some quality":
 
This annotation can be stated differently as "some anatomical entity, in a particular taxon, has some quality":
 
  '''New form'''
 
  '''New form'''

Revision as of 03:07, 4 March 2009

Our phenotype annotations currently employ linking a phenotype composition to a taxon via an exhibits relation. However this model has not provided an obvious way to employ taxon-specific homology annotations to the anatomical entities being annotated. This page describes an alternative representation of phenotype annotations which seems to provide a solution for employing homology annotations as well as anatomical relations which are taxon-specific.

Phenotype annotations

Old form

Our current phenotype annotation model has the form "a particular taxon has some phenotype, consisting of some quality inhering in some anatomical entity":

Danio rerio exhibits sigmoid^inheres_in(vertebra 1)

The phenotype is a post-composition with the following properties:

sigmoid^inheres_in(vertebra 1) is_a sigmoid
sigmoid^inheres_in(vertebra 1) inheres_in vertebra 1

New form

This annotation can be stated differently as "some anatomical entity, in a particular taxon, has some quality":

New form
vertebra 1^in_taxon(Danio rerio) has_quality sigmoid

The entity is a post-composition with the following properties:

vertebra 1^in_taxon(Danio rerio) is_a vertebra 1
vertebra 1^in_taxon(Danio rerio) in_taxon Danio rerio

Taking into account the contents of the TAO and TTO ontologies, we can infer useful statements such as:

vertebra 1^in_taxon(Danio rerio) is_a vertebra 1^in_taxon(Cyprinidae)
vertebra 1^in_taxon(Danio rerio) is_a bone

Taxonomically variable ontology relationships

Using these entity post-compositions, we can assert taxon-specific anatomical relationships such as:

vertebra 1^in_taxon(Cyprinidae) part_of Weberian apparatus

Now if we search for phenotype annotations in which the subject entity is part_of the Weberian apparatus, we will find the one about vertebra 1 of Danio rerio:

vertebra 1^in_taxon(Danio rerio) has_quality sigmoid

However, we would correctly not find an annotation about vertebra 1 in Chanos chanos, which is not in Cyprinidae:

vertebra 1^in_taxon(Chanos chanos) has_quality serrated

Searching for phenotype annotations in which the subject entity is vertebra 1 would return both statements.

Homology annotations