other: distinguish metabolism of encoded and unencoded molecules?

[gocentral]

Michale Ashburner has made a distinction between encoded molecules (directly or indirectly encoded: DNA, RNA, proteins; mostly large) and unencoded ones (the rest; mostly small) which is useful as the members of each group have functional, physiological, even chemical features in common than distinguish them from the other. There are no high level grouping terms in GO that allow this distinction and classification. "Metabolism of macromolecules" (GO:0043170) almost works for encoded entities but it includes polysaccharides; "primary metabolism" (GO:0044238) covers the metabolism covers many important classes of unencoded molecules (lipids, carbohydrates, amino acids, nucleobases) but also explicitly excludes some (nitric oxide, ubiquinone); "small molecule metabolism" (GO:0044281) picks up many of these but excludes lipids, carbohydrates, and amino acids.

Indeed, this distinction doesn't look like it fits the definitions of the two terms: Primary: The chemical reactions and pathways involving those compounds which are formed as a part of the normal anabolic and catabolic processes. These processes take place in most, if not all, cells of the organism. Small-molecule: The chemical reactions and pathways involving small molecules, any low molecular weight, monomeric, non-encoded molecule. The small-molecule term from its definition sounds more general, but as currently used neither term seems appropriate to cover all unencoded molecules.

Would it be possible, perhaps, to broaden the definition of "small-molecule" in practice to encompass all unencoded molecules, whereupon "primary metabolism" becomes a subset of it?

Reported by: deustp01

Original Ticket: geneontology/ontology-requests/10200

[gocentral]

Here's the SF ticket on the previous discussion of defining 'small molecule': https://sourceforge.net/p/geneontology/ontology-requests/8853/ We decided to exclude polymers since it would be odd to call these small molecules, especially since 'polysaccharide metabolic process ; GO:0043170' is currently under 'macromolecule metabolic process ; GO:0043170'.

Becky

Original comment by: rebeccafoulger

[gocentral]

• assigned_to: Paola Roncaglia

Original comment by: paolaroncaglia

[gocentral]

OK. The previous discussion finished without resolving the issue of whether an encoded-molecule / unencoded-molecule distinction would be useful in GO. I think it would be. As Becky notes, size alone has problems: glycogen (large) metabolism and its regulation have much in common with glucose (small) metabolism and its regulation, at least in mammals.

From the previous discussion, it looks like trying to broaden the definition of "small molecule metabolism" would cause problems (sorry for missing that), so this becomes, I guess, two new term requests, for metabolism of encoded and unencoded molecules. If the basic idea seems plausible, I'll try to compose definitions and suggest parents and children.

Peter

Original comment by: deustp01

[gocentral]

Hi Peter,

Yes, if you could please compose definitions and suggest parents and children, we can then further discuss your proposal among editors (I'll make a note for our next meeting).

Thanks, Paola

Original comment by: paolaroncaglia

[gocentral]

Set to pending - Peter to propose definitions.

Original comment by: paolaroncaglia

[gocentral]

• status: open --> pending

Original comment by: paolaroncaglia

[gocentral]

The rationale for requesting GO biological process terms for metabolism of encoded molecules and metabolism of unencoded molecules was to provide a convenient way to group metabolism of DNA, RNA, and proteins in one place in our event hierarchy and metabolism of essentially everything else in another. It looks like a definition of “encoded” as a role is possible, but it has an unintended side effect that would keep the “encoded” and “unencoded” terms from being usable for grouping as we wanted, so I’m withdrawing the term request.

An encoded molecule is one that transmits (DNA, some RNA) or embodies (all other RNA, proteins) genetic information. Replication, transcription, and translation processes could thus fit under the heading of “metabolism of encoded molecules”. Other processes in which DNA, RNA, and proteins are metabolized, however, do not fit so well, e.g., degradation of any of them, or most covalent modifications of them, which surely affect how the information in the molecule is used but don’t necessarily change the information.

As a result, at least for Reactome, if we impose a division of metabolism into that of encoded molecules and that of unencoded molecules, it looks like we’d need to separate, e.g., protein synthesis from all other aspects of protein metabolism (assisted folding, covalent modification, degradation), which we don’t want to do. Exactly as GO already does, we want to group annotations of a molecule's encoding role with ones of its other roles rather than group by role across chemical classes. If we use existing GO terms, we get a metabolism event hierarchy like this: GO:0008152 metabolic process

• GO:0043170 macromolecule metabolic process
• - GO:0006259 DNA metabolic process
• - GO:0016070 RNA metabolic process
• - GO:0019538 protein metabolic process
• GO:0005975 carbohydrate metabolic process
• GO:0006629 lipid metabolic process
• etc. This preserves the functional groupings we want consistent with the current GO metabolism is_a hierarchy. The major oddity is cosmetic: some very big processes (all of macromolecule metabolism) appear as siblings of some very limited ones (glyoxylate metabolism).

Original comment by: deustp01

[gocentral]

Thanks Peter (and sorry for the delay: somehow SF didn't notify me of your comment, ...or it did but the email fell through the cracks). So, I'm closing this ticket. If you have any comments, please feel free to re-open it.

Best, Paola

Original comment by: paolaroncaglia

[gocentral]

• status: pending --> closed-works-for-me

Original comment by: paolaroncaglia