Refine definition of 'complex of molecules' and 'protein-containing complex' to involve cardinality

[sebastianduesing]

During the 2025-06-02 COB development call, we discussed the idea that the distinction between the PRO classes 'protein-containing complex' (a subclass of 'complex of molecules') and 'protein aggregate' is that protein-containing complexes and complexes of molecules have a characteristic/defined/etc. number of proteins/molecules, such that removing one essentially changes the complex, while protein aggregates are not that well-defined; beyond a certain size, a protein aggregate is the same whether it contains n or n+1 proteins.

We proposed editing the definition of 'complex of molecules' to involve cardinality in this way, in order to be able to clearly distinguish complexes from aggregates or other material entities composed of molecules.

[nataled]

The current definition of 'complex of molecules': "A complex of two or more molecules that are not covalently bound."

Straw man for addition of cardinality as a criterion: "A complex of two or more non-covalently bound molecules for which the specific number of components is key to its distinct identity and functioning; that is, addition or removal of any component will markedly change one or more important aspects of the complex."

The current definition of 'protein-containing complex': "A stable assembly of two or more macromolecules, i.e. proteins, nucleic acids, carbohydrates or lipids, in which at least one component is a protein and the constituent parts function together."

Straw man for addition of cardinality as a criterion (paraphrasing from Wishnie et al): "A stable assembly of two or more macromolecules, i.e. proteins, nucleic acids, carbohydrates or lipids, in which at least one component is a protein, and the constituent parts function together only when all components are present within the complex."

Possible additional criterion: "...for which the specific number and arrangement of components..."

===

The following is off-topic in a strict sense (because it isn't about cardinality), but presented here because it is a fairly simple issue concerning 'complex of molecules'. Willing to break it out into a different ticket if there are objections.

The original definition of 'complex of molecules' (and the new one proposed above) have a slight air of circularity. This can be addressed by changing "A complex of two or more..." to "A stable association of two or more..."