Add homology similarity scores

[hkir-dev]

Solves issue: #280

But still most of the homologous relations don't have a similarity score (Such as https://github.com/obophenotype/brain_data_standards_ontologies/blob/issue-280/src/patterns/data/default/CCN202002013_class_homologous.tsv).

After updating centralized data following subcluster of changes occurred that need to be confirmed:

Marmoset individuals (src/templates/CCN201912131.tsv)

old : PCL:0015579	RNAseq 117 - CS201912131_79	Oligo L2-6 OPALIN MAP6D1	PCL:0015640|PCL:0015647|PCL:0015651
new : PCL:0015579	RNAseq 117 - CS201912131_79	Oligo L2-6 OPALIN MAP6D1	PCL:0015640|PCL:0015647


old : PCL:0015617	RNAseq 116 - CS201912131_117	Exc L5-6 FEZF2 LPO	PCL:0015639|PCL:0015647
new : PCL:0015617	RNAseq 116 - CS201912131_117	Exc L5-6 FEZF2 LPO	PCL:0015639

old : PCL:0015618	RNAseq 118 - CS201912131_118	OPC L1-6 PDGFRA COL20A1	PCL:0015647|PCL:0015651|PCL:0015683
new : PCL:0015618	RNAseq 118 - CS201912131_118	OPC L1-6 PDGFRA COL20A1	PCL:0015647|PCL:0015683

old : PCL:0015647	RNAseq 116-120 - CS201912131_147	Oligo-OPC	PCL:0015682
new : PCL:0015647	RNAseq 117-120 - CS201912131_147	Oligo-OPC	PCL:0015651

[shawntanzk]

@hkir-dev - dont quite understand what the last column (with the changes) are?

[hkir-dev]

It is the I 'subcluster of' SPLIT=| column. Such as:

    <owl:NamedIndividual rdf:about="http://purl.obolibrary.org/obo/PCL_0015647">
        <obo:RO_0015003 rdf:resource="http://purl.obolibrary.org/obo/PCL_0015651"/>
        ...
    </owl:NamedIndividual>

[shawntanzk]

ah ok, the changes arent wrong - seems like oligo-opc got mapped as a subcluster_of non-neuronal instead of a intermediary node (I think)

[shawntanzk]

most of the homologous relations don't have a similarity score

hmm are you using these:
https://raw.githubusercontent.com/AllenInstitute/MOp_taxonomies_ontology/main/mouseMOp_CCN202002013/Mouse_CrossSpecies_Similarity.csv
https://raw.githubusercontent.com/AllenInstitute/MOp_taxonomies_ontology/main/marmosetM1_CCN201912132/Marmoset_CrossSpecies_Similarity.csv
https://raw.githubusercontent.com/AllenInstitute/MOp_taxonomies_ontology/main/humanM1_CCN201912131/Human_CrossSpecies_Similarity.csv
cause they look like they have all the scores regardless of if there's a relation, and we want to only represent those with the relation

[hkir-dev]

Yes we are using those but they don't contain all of the nodes. Seems they only have the leaf nodes (Cell Type) in those tables. For example accession id >= CS202002013_117 is not in the mouse similarity file.

[shawntanzk]

I'll create a ticket on that on allen repo