Bioprotocols / labware-databank

MIT License
2 stars 0 forks source link

Compatability #5

Open photocyte opened 1 year ago

photocyte commented 1 year ago

Just a note on the compatibility relationship between labware-databank items. I think it should be divided into at least 3 classifications:

Sometimes you find that things aren't a perfect fit, but they still sort of work. i.e. Weakly not compatible. For example, the 290-8219-03L plate lid, is strongly incompatible with the AWLS-S30023 reservoir, but the AB-0755 is weakly incompatible with it.

photocyte commented 1 year ago

On second thought: I'm certain that "is compatible with" "is incompatible with" or equivalent terms is out here in a widely used ontology somewhere. For better interoperability, might stick with just using one of those existing terms.

rpgoldman commented 1 year ago

It would be good to have a pointer to an ontology that makes such relations available. We would have to be careful about assuming that it's going to be compatible, since it will probably have restrictions on what sorts of things can be compatible with what other sorts of things. Those restrictions may or may not be compatible with the labware-databank entities.

Even if not directly usable, though, we may be able to work around restrictions by providing translations (which might involve translating both the relation and the entities that play roles in the relation).

photocyte commented 1 year ago

I gave a quick search: https://bioportal.bioontology.org/search?q=compatible&ontologies=&include_properties=false&include_views=false&includeObsolete=false&require_definition=false&exact_match=false&categories=

My preference would be to use whichever ontology was most "upstream" & perhaps had the least restrictions on what things can be mutually compatible.

rpgoldman commented 1 year ago

I looked at those and I don't think they fit our needs. Most of them seemed to define "compatible with" as "possibly coexisting" which is weaker than what we want.

For our purposes "compatible with" is stronger: it means something like "this device will function correctly when this piece of labware is used (in some expected fashion) with it." The first definition is more like "this device won't blow up if this piece of labware is in the same space."

photocyte commented 1 year ago

Came across this recently, seems relevant (perhaps a bit off topic for this particular thread): https://github.com/biopragmatics/bioregistry

photocyte commented 1 year ago

Okay, here is an ontology term which is a bit "stronger", and potentially suitable, although I have no idea if the particularities of how it is defined will make it incompatible (as our labware, are not necessarily medical devices/products, but it's not that far off...)

https://www.ebi.ac.uk/ols4/ontologies/ncit/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FNCIT_C139494