protein processing def look like modification) GOSLIM

[gocentral]

protein processing def The posttranslational modification of a protein, particularly secretory proteins and proteins targeted for membranes or specific cellular locations.

hmmm, from this def it sounds like lots of things should be under this term but aren't.....like GPI anchor biosynthesis protein amino acid glycosylation.

But the child terms don't really fit the definition???

I spotted this because S. cerevisae pig-x (PBN1) is annotated to protein processing in addition to GPI anchor biosynthetic process

Reported by: ValWood

Original Ticket: "geneontology/ontology-requests/4343":https://sourceforge.net/p/geneontology/ontology-requests/4343

[gocentral]

Logged In: YES user_id=554736 Originator: NO

I am guessing protein processing refers to processes where a Nterminal signal sequence is clipped off or something?

In any case, I think GPI anchor synthesis is an independent process?

I haven't looked it up thoroughly. I will see what I can find.

rama

Original comment by: rbalakri

[gocentral]

Logged In: YES user_id=516865 Originator: YES

Yep there seems to be something odd with the def here.....the definition looks like modification, but the child terms are processing...

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=436423 Originator: NO

This has come up before, in SFs 1232819 and 1000275 (maybe others; 1000275 is the open one) ...

https://sourceforge.net/tracker/index.php?func=detail&aid=1000275&group\_id=36855&atid=440764

it's been a mire for ages ... anyone interested in sorting it out finally? m

Original comment by: mah11

[gocentral]

Logged In: YES user_id=516865 Originator: YES

Not me, I don't know enough about this area, and I have about ~10 big projects on my to do list which are way behind schedule.

Original comment by: ValWood

[gocentral]

assigning to Harold because he already has the older item linked in comments

Original comment by: mah11

[gocentral]

• assigned_to: nobody --> hjdrabkin

Original comment by: mah11

[gocentral]

What child terms don't fit. It doesn't HAVE to be secretory proteins; A protein phosphrorylation is protein processing also.

Original comment by: hdrabkin

[gocentral]

Hi Harold,

protein processing is defined The posttranslational modification of a protein, particularly secretory proteins and proteins targeted for membranes or specific cellular locations. HOWEVER the children are GO:0019082 viral protein processing GO:0016486 peptide hormone processing GO:0016540 protein autoprocessing GO:0006465 signal peptide processing GO:0007323 peptide pheromone maturation GO:0031638 zymogen activation GO:0007225 patched ligand processing GO:0035103 sterol regulatory element binding protein cleavage GO:0006624 vacuolar protein processing GO:0030908 protein splicing GO:0034982 mitochondrial protein processing

and appear, from the term names to refer to post-translational pepetide cleavage events. However the definition of protein -processing doe not restrict this term to these events. The way the term is currently defined it could refer to, for instance, other pt-mods like protein phosphorylation, GPI anchor attatchment ubiquitination etc.

My feeling from looking at the term, its position and its children is that 'processing' here is meant to refer to proteolytic cleavage, but that is not conveyed by the def. Possibly all that is needed is that the definition neads tightening so it refers to proteolytic cleavage?

However, I now notice this is the opposite of your comment as you are saying you would include phosphorylation as processing.

VAl

Original comment by: ValWood

[gocentral]

Something could have separate annotations to both from different papers depending on the experiment. Are the two annotations coming from the same or different papers? A note on protein processing

1. can include cleavages
2. can include modifications
3. can also be co-translational (e.g., proline hydroxylation does occur co-translationally during collagen biosyntheis; however, it can also happen after the complete chain is realeased. I think the def draws undue attention to a narrow range.

I think the def needs to be tweeked. to be more in line with 1-3 above.

Original comment by: hdrabkin

[gocentral]

Hi Harold I'm not annotating anything, I can't use this time right now as the definition is not compatible with the children. If the definition is correct, then other terms should be added as children. Otherwise the definition needs revising to narrow the scope of the term.

val

Original comment by: ValWood

[gocentral]

I didn't read that properly. You think it should be brodened (and all modification terms moved to be children)

Do you think that this is what this terms was meant to represent, or do you think it was supposed to refer only to cleavage?

Val

Original comment by: ValWood

[gocentral]

The term "processing" includes modification and cleavages. This is for both RNA and protein processing. one should have x processing is_a x cleavage is_a x modifcation

At RNA processing meetings, they have sessions on nucleases, and sessions of base-modification; I would imaging the same for protein processing meetings.

But I am still not clear why you thing the current children of protein processing are not consistent with the definition. The definition does not state cleavage only. I seem to be missing a point here.

Original comment by: hdrabkin

[gocentral]

Hi Harold,

OK, so if this term covers all of modification then the current chidren are OK, but lots of childeren (all of the other protein moddification terms, phosphoylation, ubiquitination. ned to be moved under here also. At present they are all siblings of protein processing.

at the moment we have

GO:0043687 post-translational protein modification --GO:0016485 protein processing --all of the other protein modification terms

Original comment by: ValWood

[gocentral]

I took a good look at the graph, and it appears that the mods were all set up parallel with the"processing term" as children of PTM SO, perhaps the less damage to the tree is to make the processing term explicitly state cleavage Both are then under post-trans modifications PTM). I would think PTM should have a processing term as a synonym for people like me who think processing = cleavage and mod as opposed to people who thing ptm = processing (cleavage) + mod I think that making the def explicit will solve the problem and making another synonym for PTM will help.

I would go ahead and use the processing as meaning cleavage as it appears this is what the orginator of the term had in mind. If no disagreement, I can change the def to say something like .The posttranslational cleavage of a protein, particularly secretory proteins and proteins targeted for membranes or specific cellular locations."

Original comment by: hdrabkin

[gocentral]

But now, I just realized this is not working if processing is cleavage only:

GO term: post-translational protein modification Synonym: PTM GO id: GO:0043687 Definition: The covalent alteration of one or more amino acids occurring in a protein after the protein has been completely translated and released from the ribosome.

The def would need to be changed completely, as cleavage is not alteration of one or more amino acids. nuts

We need to make a new single processing term to describe post-translational alteration of a polypeptide after the protein has been completely translated and released from the ribosome. (Protein alteration)

Then, put current PTM as child Pull current processing term out from under PTM and make sib of TPM

That should do it with least amt of damage.

Original comment by: hdrabkin

[gocentral]

once more The term "GO:0006464 protein modification process would serve as the parent IF we remove GO:0019988 charged-tRNA amino acid modification out from under it because this is happening NOT at the protein level but at the aminoacyl-tRNA level. charged-tRNA amino acid modification is a type of biopolymer modification THEN, we move processing to be a sib of ptm The def onf GO:0006464 would need to be changed to reflect any alteration of the original sequence and to remove the part about the aminoacyl-tRNA, which is not a part of any protein modification. It's to make the amino acyl-tRNA needed for protein synthesis when the organism does not have a separate synthetase to make it from the amino acid directly. I have attached proposed structure as pdf pic.

Original comment by: hdrabkin

[gocentral]

Proposed reorg of protein modification

Original comment by: hdrabkin

[gocentral]

Hi Harold,

I see what my problem is now. Looking at all of the annotations to "protein processing" they seem to encompass changes involved in protein maturation (so it would exclude most, but possibly not all, occurances of specific modifications, for instance reversible methylations, phosphorylations etc would not be considered to be protein processing events)

The current positioning is probably OK, but the def should say that processing is modification resulting in a mature protein. At the moment the def is indistinguishable from the protein modification term.

If, and when annotations to protein processing are required for other modifications,

child terms should be added,

i.e. protein processing by xxx where protein processing by xxx would also be a child of the modification xxx

Val

Original comment by: ValWood

[gocentral]

Hi Val

It's a mess for a numer of reasons anyways.. sigh What do you think of my proposed change (see pdf); it doesn't require additional terms and clearly separates chain modification from chain cleavage. I'm hoping Midori or Jane will take a look; no matter what we do some defs need to be altered.
(note the parent to all (protein modification) says "e covalent alteration of one or more amino acids occurring in proteins, " this would automatically preclude "cleavages", yet protein processing is a child already, so it's still wrong, unless one wants to say that breaking a bond between two amino acids is a covalent alteration of the amino acid. I don't think so. It''s actually an alteration of the peptide bond. (sorry my responses were a bit fragmented yesterday... working on my laptop not attached to the dock and laptop keyboard needs to be replaced). Harold

Original comment by: hdrabkin

[gocentral]

I'm not quite sure I understand the pdf, but it seems to me that we could either do this:

protein processing GO:0016485 [i] protein cleavage (GO:new, also is_a proteolysis) [i] protein modification GO:0006464 [i] post-translational protein modification GO:0043687

or this:

protein modification GO:0006464 [i] protein processing GO:0016485 [i] post-translational protein modification GO:0043687

Either way, I agree about moving GO:0019988 charged tRNA yadayada.

The big problem is that the existing defs (and previous rounds of inconclusive discussion) don't give any good indication of which arrangement may have been preferred way back when. The 'protein processing' def is particularly useless -- it does NOT say that it's meant to refer only to cleavage (it hardly says anything at all), so I think Val's concern about broadening it is unfounded.

In fact, I prefer the interpretation that Harold seems (from the 2009-01-22 16:20 comment) to favor: we can broaden the definition of the existing 'protein processing' term to include both cleavage and covalent modification, as in the first snippet above. I like this better because it's always less risky to broaden a definition than narrow it -- if we narrow a def, there's the risk that some annotations will be thrown off.

If we broaden the protein processing def, protein modification would become its child, so the terms now under modification would acquire a path to processing, addressing Val's point from the 2009-01-22 16:03 comment.

For protein maturation, the ontologically correct approach is to add a term for 'protein processing involved in protein maturation'; the existing term 'protein maturation via proteolysis' would be related somehow, probably as an is_a child.

I'm rashly assuming that we'll come up with nicer definition wordings once we decide on the high-level structure.

m

Original comment by: mah11

[gocentral]

Hi Midori

Protein modification can't go under protein processing because lots of PT modifications (some phosphorylations, ubiquitinations etc) are not necessarily processing. I don't think that reversible phophorylations etc would be considered processing would they?

The other way around (as it is now) seems OK, but the definition of protein processing does not adequately distinguish it from modification (I think that curently I could annotate to these two terms interchangeably). This indicates to me that there is something missing from the definition of 'protein processing', that part of the process which distinguishes it from reversible phophorylations etc.

val

Original comment by: ValWood

[gocentral]

Hi Midori: The pdf was trying to out line changes needed if we went up one more level from the direct parent of the processing term, because it too refered only to changes of an amino acid in a chain (thus negating cleavage).

I'll see if I can make it more clear; I have also made an obo version with the proposed change, and will attempt to add as a compressed file

Original comment by: hdrabkin

[gocentral]

Val - the point is, it depends on how "processing" is defined. It's clear from this and previous rounds of hair-pulling that there's more than one usage out there in the wild. If we don't go with the broader meaning of "processing" that Harold suggested earlier -- which would encompass all modifications, maturation or otherwise, reversible or not -- then we do need to figure out not only what the difference between processing and modification is, but also how they should be related. Processing does include cleavage, and I think we've agreed, or at least tended to think, that modification doesn't.

Original comment by: mah11

[gocentral]

Harold has just emailed that he's now leaning toward restricting processing to cleavage. I'll let him fill in more details, since in this area he knows what he's talking about and it could be argued that I don't ;)

Original comment by: mah11

[gocentral]

OK...I don't think I have come across all of modification described as processing, just thos modification events which result in the formation of a mature protein product. I wonder if we should avoid processing as a primary term name anyway, Its clearly ambiguous and I don't know if its really universally used and whatever it means could be better otherwise described...

I'll keep watching this space anyway

val

Original comment by: ValWood

[gocentral]

Here is summary of current situation from bottom up, then top down

Currently Protein Processing is a child of post-translational protein modification which is a child of protein modification process.

Current defs are

GO:0016485 protein processing: The post-translational modification of a protein, particularly secretory proteins and proteins targeted for membranes or specific cellular locations.

GO:0043687 post-translational protein modification :The covalent alteration of one or more amino acids occurring in a protein after the protein has been completely translated and released from the ribosome.

GO:0006464 protein modification: The covalent alteration of one or more amino acids occurring in proteins, peptides and nascent polypeptides (co-translational, post-translational modifications). Includes the modification of charged tRNAs that are destined to occur in a protein (pre-translation modification). ( The last sentence should be removed and “charged-tRNA amino acid modification” moved out from under protein modification)

SO, with current def of 6464 :looks more like modification. I suggest changing def of 6464 to read “the covalent alteration of a protein, peptides, or nascent polypep by cleavage or amino acid modification.

Then leave 43687 as is

Then move 6485 out from 43687 but under 6464, and change 6485 (processing) to read protein cleavage: the modfication of a protein sequence by cleavage of peptide bonds. This includes removal of pre-sequences, etc.

So new hierarchy becomes

GO:0006464 <<< changed def .GO:0043687 ..GO:0016485 <<<< changed def

We should add synonyms where appropriate.

There is tenuous distinction between whether a modification occurs co-translationally (proline hydroxylation), or post (phosphorylations), etc. But let's worry about that later.

Original comment by: hdrabkin

[gocentral]

argh I put the period in the wrong place; 43687 and 16485 are sibs, and both children of 6464 ; sorry, that's what I get for coming in with a cold bug!

GO:0006464 <<< changed def (includes everything) .GO:0043687: changes to an amino acid in the potein .GO:0016485 <<<< changed def, cleavage of the chain in any way

Original comment by: hdrabkin

[gocentral]

OK...I don't think I have come across all of modification described as processing, just thos modification events which result in the formation of a mature protein product. I wonder if we should avoid processing as a primary term name anyway, Its clearly ambiguous and I don't know if its really universally used and whatever it means could be better otherwise described...

I'll keep watching this space anyway

val

Original comment by: ValWood

[gocentral]

I just had a chat with John Garavelli, and he agrees with the proposal for

GO:0006464 protein modification <<< changed def (includes everything) .GO:0043687 post-translational mod : changes to an amino acid in the potein .GO:0016485 protein processing (but see below - possible name change)

He also recommends making the terms for specific types of modification siblings instead of children of GO:0043687, because many of them can be either co- or post-translational (often depending on what species it's in).

One other thing John suggested is renaming GO:0016485 to 'peptide chain processing' or 'protein peptide chain processing', to make it clearer that it means changes to the peptide bonds. We should also make sure the definition says this. He would include some cases of peptide bond formation as well as cleavage, such as protein splicing or cyclization. He says it would be worth changing the name of 'protein autoprocessing' similarly, but not to bother with the other child term names. If you don't like the idea of renaming, we could add synonyms instead.

I'm just the messenger, but let me know if you have questions to pass on ... m

Original comment by: mah11

[gocentral]

These things sound sensible.

It adresses my main concern which was that the definiition of processing is contains "Modification" and they are not under "protein modification"

so yes

Original comment by: ValWood

[gocentral]

this one sounds fairly straightforward, can this one be done, or are we still waiting on something? val

Original comment by: ValWood

[gocentral]

This one has cropped up again whils slim building

We noticed a number pf potential problems

mainly that protein processing is defined as: "the post translational modification"

but is not a child of protein modification?

and the unclear distinction between processing and maturation

this also seems odd protein maturation --protein processing ---protein maturation by peptide bond cleavage,

especially when you look at the siblings.

I'm putting this one up for the clinic on 26th

Original comment by: ValWood

[gocentral]

• summary: protein processing def looks wrong (like modification) --> protein processing def look like modification) GOSLIM

Original comment by: ValWood

[gocentral]

This one has cropped up again whils slim building

We noticed a number pf potential problems

mainly that protein processing is defined as: "the post translational modification"

but is not a child of protein modification?

and the unclear distinction between processing and maturation

this also seems odd protein maturation --protein processing ---protein maturation by peptide bond cleavage,

especially when you look at the siblings.

I'm putting this one up for the clinic on 26th

Original comment by: ValWood

[gocentral]

• status: open --> closed

Original comment by: ValWood

[gocentral]

• milestone: --> PomBase

Original comment by: mah11