search

ebi-chebi / ChEBI

Chemical Entities of Biological Interest (ChEBI) is a freely available dictionary of molecular entities focused on ‘small’ chemical compounds.
https://www.ebi.ac.uk/chebi
Creative Commons Attribution 4.0 International
42 stars 10 forks source link

"calcium ion" is not "metal cation" #4530

Open jpquast opened 3 weeks ago

jpquast commented 3 weeks ago

Technically CHEBI:39124 calcium ion should not have the outgoing ontology to CHEBI:25213 metal cation since it is an ion and not a cation. "calcium cation" is correctly connected to "metal cation" through chebi:33513 alkaline earth cation. So ideally the outgoing ontology of "calcium ion" to "metal cation" should be removed.

cmungall commented 3 weeks ago

This is an example of the cryptic equivalence anti-pattern. A lot of ontologies had analogous errors before they adopted logical axiomatization. I am still hopeful that one day CHEBI will adopt logical axiomatization, this kind of thing becomes very easy to detect automatically.

In this case, I believe CHEBI are kind-of-sort-of saying that calcium ion and cation are equivalent because the anion form would be so unstable, hence all calcium ions are cations?

See:

image

But if this is the intent, I think it's better to be explicit, and either:

  1. merge the two classes, making the ion term a broad synonym, and adding a comment about unlikeliness of the state
  2. leave open the possibility of anion states, and modify the structure to be coherent as @jpquast suggests, adding a note that the broader ion term only exists for people using CHEBI for theoretical chemistry purposes
jpquast commented 3 weeks ago

Thanks for adding to this @cmungall ! In case of calcium the anions would likely be indeed unstable, but for some other metals this might not be the case (e.g. gold, CHEBI:30030). So I think if the term is "metal cation" the linked atomic ions should also be the "cations". Also calcium is the only atom for which the ion is linked to the "metal cation" term and not the cation.