add KEGG disease ids about monarch-disease-ontology-retired HOT 13 CLOSED

From @nlwashington on March 6, 2015 17:34

note that these are different than the "disease pathways".

from monarch-disease-ontology-retired.

From @nlwashington on April 21, 2015 18:24

there's definitely some oddities. for example:
http://www.kegg.jp/dbget-bin/www_bget?ds:H00493

this is clearly a grouping class, as it is named:
Heparan sulfate proteoglycan gene defects, including:
Dyssegmental dysplasia, Silverman-Handmaker type;
Schwartz-Jampel syndrome;
Simpson-Golabi-Behmel syndrome, type 1;
Omodysplasia 1;
Multiple exostoses

which includes 5 diseases, that aren't themselves identified with kegg ids. the genes for the specific sub-types are mapped to the grouping class rather than the specific disease, which is really unfortunate.

from monarch-disease-ontology-retired.

Just FYI the LCS of those 6 OMIMs in 'bone disease'

from monarch-disease-ontology-retired.

From @nlwashington on April 21, 2015 23:21

i notice only a couple of the diseases are linked to something with "perlecan" in it. I didn't know what that was, so i looked:
Perlecan (PLC) also known as basement membrane-specific heparan sulfate proteoglycan core protein (HSPG) or heparan sulfate proteoglycan 2 (HSPG2), is a protein that in humans is encoded by the HSPG2 gene.

I wonder if there are missing links?

but it looks like the LCS is actually just "disease", though most are LCS with "bone disease". Simpson-Golabl-Behmel is missing out ?

from monarch-disease-ontology-retired.

Ah yes you're right, Simpson-Golabl-Behmel is effectively unclassified

from monarch-disease-ontology-retired.

From @nlwashington on May 27, 2015 16:3

what is the URI for kegg disease identifiers? for example:
http://www.kegg.jp/dbget-bin/www_bget?ds:H00060

in dipper, i could add equivalence axioms between kegg disease ids and omim disease ids if they are 1:1. this is a very conservative estimate. would you like that?

from monarch-disease-ontology-retired.

From @nlwashington on May 27, 2015 18:41

there are also ~73 KEGG diseases that have no omim mappings, but do have disease-gene associations. for example:
http://www.kegg.jp/dbget-bin/www_bget?ds:H00014

be sure you get these into mondo!

from monarch-disease-ontology-retired.

ttl file: http://nif-crawler.neuinfo.org/monarch/ttl/kegg.ttl

Ok so seems we have some options here:

1. we could create disease classes for each of the diseases in the KEGG file, and then create equivalencies to things in MONDO
2. we could add xrefs for each KEGG disease to something in MONDO, adding DC classes where needed.
   (presumably this is what you did @cmungall, but maybe rethink this?)

from monarch-disease-ontology-retired.

I think we will need to create the equiv classes. Lets run obol against MONDO, review potential equivalencies, existing ones, and create DC classes where we need them.

from monarch-disease-ontology-retired.

nevermind - lets do this instead- grab all set of disease-gene associations that only come from KEGG, and then we'll curate these separately, its just not that many. That way we don't need to ruse KEGG data. please also dump the papers associated those data.

from monarch-disease-ontology-retired.

@cmungall curious what the status is on adding kegg disease IDs? Let me know if you need something from the dipper ingest. Also related #146, https://github.com/monarch-initiative/monarch-app/issues/1514

from monarch-disease-ontology-retired.

KEGG equivalencies are in mondo. What URIs should we use for them?

from monarch-disease-ontology-retired.

This issue was moved to monarch-initiative/mondo#167

from monarch-disease-ontology-retired.