Add ncbitaxon:has_division property.

[hrshdhgd]

This partially addresses #132

I have made minor tweaks to the code where I capture division properties from NCBITaxon's division.dmp file and add the property to all relevant classes.

Homo Sapiens class now looks like this

<!-- http://purl.obolibrary.org/obo/NCBITaxon_9606 -->

    <owl:Class rdf:about="http://purl.obolibrary.org/obo/NCBITaxon_9606">
        <rdfs:subClassOf rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_9605"/>
        <ncbitaxon:has_division rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_Primates"/>
        <ncbitaxon:has_rank rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_species"/>
        <oboInOwl:hasDbXref>GC_ID:1</oboInOwl:hasDbXref>
        <oboInOwl:hasExactSynonym>human</oboInOwl:hasExactSynonym>
        <oboInOwl:hasOBONamespace>ncbi_taxonomy</oboInOwl:hasOBONamespace>
        <rdfs:label>Homo sapiens</rdfs:label>
    </owl:Class>
    <owl:Axiom>
        <owl:annotatedSource rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_9606"/>
        <owl:annotatedProperty rdf:resource="http://www.geneontology.org/formats/oboInOwl#hasExactSynonym"/>
        <owl:annotatedTarget>human</owl:annotatedTarget>
        <oboInOwl:hasSynonymType rdf:resource="http://purl.obolibrary.org/obo/ncbitaxon#genbank_common_name"/>
    </owl:Axiom>

The one doubt I have is should we capture this information as a string or I assigned them URIs (like it is done with has_rank). Open to comments. If this is not acceptable, it's ok. Please let me know.

[matentzn]

Does divison have to be a PURL? it seems adding PURLS like http://purl.obolibrary.org/obo/NCBITaxon_Primates will create a lot of weird identifiers..

[hrshdhgd]

I'm actually ok with strings. I agree PURL would look weird. I wasn't sure. I'll roll it back for now.

[hrshdhgd]

Rolled back, now Homo Sapiens looks like this:

<!-- http://purl.obolibrary.org/obo/NCBITaxon_9606 -->

    <owl:Class rdf:about="http://purl.obolibrary.org/obo/NCBITaxon_9606">
        <rdfs:subClassOf rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_9605"/>
        <ncbitaxon:has_division>Primates</ncbitaxon:has_division>
        <ncbitaxon:has_rank rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_species"/>
        <oboInOwl:hasDbXref>GC_ID:1</oboInOwl:hasDbXref>
        <oboInOwl:hasExactSynonym>human</oboInOwl:hasExactSynonym>
        <oboInOwl:hasOBONamespace>ncbi_taxonomy</oboInOwl:hasOBONamespace>
        <rdfs:label>Homo sapiens</rdfs:label>
    </owl:Class>
    <owl:Axiom>
        <owl:annotatedSource rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_9606"/>
        <owl:annotatedProperty rdf:resource="http://www.geneontology.org/formats/oboInOwl#hasExactSynonym"/>
        <owl:annotatedTarget>human</owl:annotatedTarget>
        <oboInOwl:hasSynonymType rdf:resource="http://purl.obolibrary.org/obo/ncbitaxon#genbank_common_name"/>
    </owl:Axiom>
    ```

[matentzn]

@hrshdhgd I think I am missing some context from other peoples work..

@cthoyt @jamesamcl @anitacaron can someone remind what the deal is with these PURLs?

        <ncbitaxon:has_rank rdf:resource="http://purl.obolibrary.org/obo/NCBITaxon_species"/>

Maybe we agreed this is indeed the right thing to do (sorry @hrshdhgd).

[jamesaoverton]

Those non-numeric IDs are prehistoric, and we should not mint new ones going forward.

[hrshdhgd]

Those non-numeric IDs are prehistoric, and we should not mint new ones going forward.

I agree, that's why I was confused to see the purls for has_rank values. But I see in #120 that TAXRANKS:XXX is proposed to be implemented which I like as an idea. Having said that, would it make sense to tie has_division to pre-existing standards or create a new one?

[anitacaron]

Is this Primates the same as this one? If so, we can use the same url then.