Oort Use Case Neural Crest

Oort Use Case: Neural Crest

Introduction

Goal: create a neural crest ontology/knowledge base by automatic extraction from existing knowledge sources.

Some of the ontologies (e.g. CL) require some work to fix their representation of NC cells, so this will be an iterative process.

Will be presented at the February NC workshop in NC

Cell types

• cl.owl

The core cell type is "neural crest cell"

      is_a CL:0000333 ! neural crest cell *** 
       is_a CL:0000008 ! cranial neural crest cell
       is_a CL:0000011 ! trunk neural crest cell
       is_a CL:0002607 ! enteric neural crest cell

We will "extend out" from here. I.e. the input DL query will be

 'neural crest cell'
  or
  'developmentally related to' some 'neural crest cell'
  or
  'neural crest'
  or
 'developmentally related to' some 'neural crest'
  or
  'has part' some 'neural crest cell'
  ...

We will explore various queries, using property chains to simplify this

Gross anatomy

• uberon/composite-vertebrate.owl

This combines Uberon with MA, FMA, CL etc

GO

• go.owl

We will take the subset of GO that references NC cells or the NC, e.g:

    po GO:0048513 ! organ development
      po GO:0060485 ! mesenchyme development
       po GO:0048762 ! mesenchymal cell differentiation
        po GO:0014031 ! mesenchymal cell development
         po GO:0001837 ! epithelial to mesenchymal transition
          is_a GO:0014029 ! neural crest formation *** 
   

we will also pull in functional annotations to these classes.

phenotypes

We will use mp.owl and hp.owl

E.g.

    is_a MP:0003091 ! abnormal cell migration
     is_a MP:0002950 ! abnormal neural crest cell migration *** 
      is_a MP:0009784 ! abnormal melanoblast migration

Connecting the ontologies

• mp/equivalence-axioms.owl
• go/x-anatomy.owl

(more to be added)

Genomic element ontologies and annotation ontologies

We will include both GO, MP and HP annotations, translated to OBO/OWL as well as representation of these genes and associated information (families etc)

Most of the translations of genes will be made available here:

• http://purl.obolibrary.org/obo/omeo/

(URL may change - ensembl reflects the fact that it came from biomart)

GAFs can be translated to OWL using OWLTools. We will use

• vertebrate GAF
• Others?

Pathway Data

We can bring in pathway data from Reactome etc. BioPAX3 has to be converted into something more obolibrary-compliant. CJM has code for this, will fill in details later.

TODO - identify pathway models

• WP2064 neural crest differentiation (human)
• Reactome?

Example Query Ontology

Prefix: obo: <http://purl.obolibrary.org/obo/>

Ontology: <http://purl.obolibrary.org/obo/uberon/queries/ncc_query.owl>

Import: <http://purl.obolibrary.org/obo/ro.owl>
Import: <http://purl.obolibrary.org/obo/uberon/merged.owl>
Import: <http://purl.obolibrary.org/obo/go.owl>
Import: <http://purl.obolibrary.org/obo/go/extensions/x-anatomy.owl>

ObjectProperty: obo:RO_0002202

    
ObjectProperty: obo:BFO_0000051

    
Class: obo:UBERON_0007844

    
Class: obo:CL_0000333

    
Class: obo:QUERY_1

    EquivalentTo: 
        obo:UBERON_0001474
         or obo:UBERON_0007844,
        obo:CL_0000333
         or (obo:BFO_0000051 some obo:CL_0000333)
         or (obo:RO_0002202 some obo:CL_0000333)
    
    
Class: obo:UBERON_0001474

    

Query Ontology in OORT

This assumes, that you already create the query ontology file (zfa.owl), have the id (e.g., http://purl.obolibrary.org/obo/QUERY_1) or name of the class containing the query, and if necessary a catalog.xml file.

1. Go to the Input/Output tab

2. Specify, which ontologies you want to be pulled in. For example, the above query would require both the purls for uberon and CL. If you do not specify these, they the will remain as import and are not materialized.

3. Go to the Query tab

4. Tick check box: 'Create Ontology from Query'

5. Select query ontology file (e.g, zfa.owl) using the select file button

6. Input query term IRI (e.g., http://purl.obolibrary.org/obo/QUERY_1)

7. (If catalog.xml is required) Switch to Advanced tab, scroll down until you see the catalog.xml file input field: set catalog.xml file using 'select file'-button

8. The Advanced tab also contains the reasoner selection

9. Switch to the Input/Output tab

10. Add the ontology files (via url or file) which should be mireoted into the new ontology

11. Check output folder, by default it will put the files in the 'OORT' folder in your home directory.

12. (Optional) You can save (and load) this or any other OORT configuration using the File Menu at the top right.

13. Run OORT by clicking on the 'make release'-button.