The RNA Ontology RNAO

1
The RNA OntologyRNAO

• Colin Batchelor
• Neocles Leontis
• May 2009

Eckart, Colin and Jane In Cambridge
2
RELATION TO TIME GRANULARITY CONTINUANT CONTINUANT CONTINUANT CONTINUANT OCCURRENT
RELATION TO TIME GRANULARITY INDEPENDENT INDEPENDENT DEPENDENT DEPENDENT
ORGAN AND ORGANISM Organism (NCBI Taxonomy) Anatomical Entity (FMA, CARO) Organ Function (FMP, CPRO) Phenotypic Quality(PaTO) Biological Process (GO)
CELL AND CELLULAR COMPONENT Cell (CL) Cellular Component (FMA, GO) Cellular Function (GO) Phenotypic Quality(PaTO) Biological Process (GO)
MOLECULE Molecule (ChEBI, SO, RNAO, PrO) Molecule (ChEBI, SO, RNAO, PrO) Molecular Function (GO) Molecular Function (GO) Molecular Process (GO)
OBO Foundry http//obofoundry.org
2
Smith et al. Nature Biotechnology 2008
3
Relationship of RNAO and neighboring ontologies

• ChEBI (Chemical Entities of Biological Interest)
• Sequence Ontology
• Gene Ontology

RNAO
ChEBI
Sequence Ontology
Relations Ontology Basic Formal Ontology
4
RnaO and OBO

• Commitment to common upper ontology (BFO)
• Commitment to dividing responsibilities between
  ontologies
• ChEBI for molecular parts (and relations?)
• SO for sequences
• GO for RNA function
• Leaving RnaO to concentrate on RNA

5
Aim

• to create a common, dynamic, controlled
  vocabulary (the RO), pertaining to RNA function
  and based on RNA sequences, secondary and
  three-dimensional structures. Thus, the central
  objective of the ROC is to identify all RNA
  features, interactions and motifs mentioned in
  the literature or appearing in databases, to
  agree upon a definition for them and to write
  that definition down in a structured manner. This
  is very timely as knowledge about RNA accumulates
  and progresses rapidly. The purposes for creating
  the RO are therefore (1) to integrate sequence
  and structural databases and (2) to create
  powerful software tools that bring advanced
  computational methods to the bench scientist.

6
RNAO Working Group meeting in Cambridge, UK,
January 2009

• Presentations on
• RNAO in OWL (Colin Batchelor)
• Conformations (Jane Richardson)
• reasoning (Chris Mungall)
• formal ontology (Thomas Bittner)
• SO (Karen Eilbeck)

• ChEBI (Kirill Degtyarenko)
• alignment and phylogeny (Rob Knight)
• Rfam (Alex Bateman)
• 2D to 3D structure (Alain Laederach)

7
Existing neighbouring ontologies

• Sequence Ontology
• (5 base pair terms, 3 pseudoknots, 2 stem loops,
  10 motif terms)
• ChEBI
• (4 RNA terms)
• Gene Ontology
• (functions performed by ncRNAs)

8
RNAO v1.0

• Implemented in OWL
• (easier to write complicated logical definitions
  in Protégé than in OBOEdit but less powerful than
  FOL and names and really not suitable for
  human-readable definitions and synonyms)
• gt100 classes
• gt20 relationships

9
Simple use cases

• Classify base pairs according to the
  Leontis-Westhof scheme (easy)
• Identify motifs in 3d structures (hard going on
  impossible)
• (Vaguer, but really the most important) Provide
  bridge between different representations

10
Base pair classification (1)

• First attempt use relations to express pairing,
  bonding and so forth (Dumontierian approach)
• Base pairing
• pairsWith, pairsWithWW, pairsWithCWW
• Backbone bonding
• fivePrimeTo, threePrimeTo, etc.

11
Base pair classification (2)

• Then (OWL Manchester syntax)
• Family1BasePair hasPart some Nucleobase and
  (pairsWithCWW some Nucleobase)
• (this works and classifies base pairings
  satisfactorily within the LW scheme)

12
(No Transcript)
13
Drawbacks

• OWL cannot handle motifsthey require a larger
  subset of first-order logic
• But RNA experts differ on definitions of motifs
  anyway!
• RNAO.owl hides the interactions behind a large
  set of relations
• Relations are hard to maintain in Protégé/OWL

14
Next steps

• Integrate with real RNA data (NDB, FR3D)
• Collect RNA motif definitions
• Formalize what we are actually hiding behind the
  relations
• (WC edge, etc. are parts of nucleotides and have
  a topological relation to other edges)
• Base pairs are parts of RNA molecules
• Basebase, basephosphate and stacking
  interactions should be treated in the same way as
  we handle covalent bonds