search

RobokopU24 / ORION

Code that parses datasets from various sources and converts them to load graph databases.
MIT License
12 stars 13 forks source link

hmdb edges pointing the wrong direction #196

Open cbizon opened 9 months ago

cbizon commented 9 months ago

Match (c:biolink:PhenotypicFeature {id:"HP:0002017"})-[a]-(x)-[b]-(d:biolink:ChemicalEntity {id:"PUBCHEM.COMPOUND:3776"}) return *

Shows a number of cases where x is a gene and there is an edge saying that the gene affects the abundance of d (the chemical). The edge is coming from HMDB.

image

This is almost certainly turned around the wrong way.

cbizon commented 6 months ago

See also https://github.com/NCATSTranslator/Feedback/issues/749.

I'm not sure that this is related. And looking back at this ticket, I'm not sure why I was convinced that the abundance edge was pointed the wrong direction. It could go either way, really. But I think we need to review the chemical/gene directions in HMDB overall.

EvanDietzMorris commented 6 months ago

These edges are pointed in the other direction now. We did previously switch the direction of the affects_abundance_of edges too though, so we should double check that was correct.

eKathleenCarter commented 5 months ago

For the affects_abundance_of edges I agree with Chris's above comment, it could go both ways. I am not sure how we would clarify which way the specific edge is pointing.

cbizon commented 5 months ago

I think it's a question of going back to HMDB and reviewing whatever documentation they have about what their tables mean, and making sure that it's consistent with how we represent the dat.