remove "amino acid activation" parent to "amino acid translation"

[Antonialock]

should

GO:0006426 glycyl-tRNA aminoacylation & GO:0006544 glycine metabolic process

be related?

[hdrabkin]

No I don't think so. We don't do this for the other tRNA aminoacylations. It's the old question of where we draw the line for substrates. Aminoacyl-trna aminoacylation are also not related to ATP metabolic process. That being said, you can work your way up to cellular amino acid metabolism.

[Antonialock]

ok cheers

[ValWood]

I think this is a bit strange though.

If "cellular amino acid metabolism" is a parent, then it must be logically correct to have the individual amino acid parents?

Although I had been wondering during "Matrix" work if it was strictly correct to describe "aminoacylation" and "amino acid metabolism"

[hdrabkin]

There is a slippery slope about this: do we want to describe everything that an amino acid is involved in as an amino acid metabolism. In this case the amino acid is not really catabolized; (the breakdown of complex molecules in living organisms to form simpler ones, together with the release of energy; destructive metabolism) GO metabolism "The chemical reactions and pathways, including anabolism and catabolism, by which living organisms transform chemical substances." Anabolism: the synthesis of complex molecules in living organisms from simpler ones together with the storage of energy; constructive metabolism. Protein synthesis kind of fits here; I'm not sure if people generally think of the role of storage of energy looking at a protein, protein is generally not used as an energy source unless the body has no other option. Not sure what you mean by if it was strictly correct to describe "aminoacylation" and "amino acid metabolism" DId you mean as' amino acid metabolism'??

[deustp01]

Jumping in here in favor of narrowly drawn process boundaries, with Harold: translation is not a kind of amino acid metabolism, and if tRNA aminoacylation belongs anywhere, I'd think it belongs there. That's also an argument, responding to Val's comment from two days ago, that tRNA aminoacylation should not be a child of cellular amino acid metabolism.

Here's where terms like "causally upstream of" stop being euphemisms for "we don't exactly know what's going on" and have real logical power. Absent amino acid synthesis or uptake mediated by process one, aminoacylation cannot happen. The two distinct, non-overlapping processes are related in a precise way.

[ValWood]

There is a slippery slope about this: do we want to describe everything that an amino acid is involved in as an amino acid metabolism. In this case the amino acid is not really catabolized; (the breakdown of complex molecules in living organisms to form simpler ones, together with the release of energy; destructive metabolism)

@hdrabkin I think we are agreeing. At present we descirbe aminoacylation as "amino acid metabolism". It has dual parentage to "amino acid metabolism" and translation" I don't think we should do this?

The "matrix" intersection between "amino acid metabolism" and "translation" is purely from "aminoacyl tRNA synthases". I think these should be annotated to "translation" only?

(to interpret this, the same terms are across the top, I only have the y-axis labelled)

I think this is what @deustp01 is saying too?

so action is to remove "amino acid activation" as a child of "amino acid metabolic process".

(I was only pointing out that in the current arrangement, the additions @Antonialock would be logically correct, and if the current arrangement persists they should be added)

[deustp01]

Yes, that's what I'm saying and so the removed parentage sounds right.

[ValWood]

I changed the title to reflect the action required.

[hdrabkin]

The aminaocylation of tRNAs is not always a part of translation. That is why we have the term aminoacylation involved in translation. For example aminoacyl-tRNAS involved in post-translational amino acid transferases glycyl-tRNA involved in some cell wall biosynthesis. So, along those lines, the aminoacyl-activation process does not always yield a tRNA that is involved in translation. However, just looking at your diagram, does that leave amino acid activation with no parent? I'm wondering if maybe a has_part relationship (that is SOME amino acid activation is involved in translation, but not all;

[hdrabkin]

Wait; was reading the diagram in the wrong direction; COFFEE so "so action is to remove "amino acid activation" as a child of "amino acid metabolic process"."

[ValWood]

yes you got it, COFFEE first!

[ValWood]

...although you are quite correct that "amino acid activation" will have no parent. Although, is this term even needed?

[ValWood]

amino acid activation is defined The modification of an amino acid to an active form, for incorporation into a peptide, protein or other macromolecule.

so it seems to be the same as GO:0006418 tRNA aminoacylation for protein translation

and will cause TPV's for the non peptide synthesis aminoacylation events?

[ValWood]

amino acid activation has NO direct annotations (any evidence code), so you could obsolete it, and then add it as a synonym to the appropriate term.

[hdrabkin]

Painfully searching for any evidence of aminoacyl-tRNA synthetase independent amino acid activation.Can't find....

[hdrabkin]

Note that tRNA aminoacylation would be the term to merge amino acid activation into. So if sever link between amino acid activation and amino acid metabolism, then merge the term into tRNA aminoacylation, I think we are good? no TPV

[ValWood]

yep.

[hdrabkin]

Found; not all aminoacid activation is tRNA dependent. aminoacyl-CoAs FEBS Lett. 2011 Mar 23;585(6):893-8. doi: 10.1016/j.febslet.2011.02.018. Epub 2011 Feb 18. Aminoacyl-coenzyme A synthesis catalyzed by a CoA ligase from Penicillium chrysogenum. Koetsier MJ(1), Jekel PA, Wijma HJ, Bovenberg RA, Janssen DB.

Coenzyme A ligases play an important role in metabolism by catalyzing the activation of carboxylic acids. In this study we describe the synthesis of aminoacyl-coenzyme As (CoAs) catalyzed by a CoA ligase from Penicillium chrysogenum. The enzyme accepted medium-chain length fatty acids as the best substrates, but the proteinogenic amino acids L-phenylalanine and L-tyrosine, as well as the non-proteinogenic amino acids D-phenylalanine, D-tyrosine and (R)- and (S)-β-phenylalanine were also accepted. Of these amino acids, the highest activity was found for (R)-β-phenylalanine, forming (R)-β-phenylalanyl-CoA. Homology modeling suggested that alanine 312 is part of the active site cavity, and mutagenesis (A312G) yielded a variant that has an enhanced catalytic efficiency with β-phenylalanines and D-α-phenylalanine. DOI: 10.1016/j.febslet.2011.02.018 PMID: 21334330 [Indexed for MEDLINE]

[ValWood]

But this isn't "tRNA aminoacylation?"

[hdrabkin]

Yes, but I cannot then merge amino acid activation into it, because there are other was to activate an aminoacid that does not involve tRNA aminoacylation. You can form aa-CoA as well as aa-TRNA; The term amin acid activation does not specifiy which (The modification of an amino acid to an active form, for incorporation into a peptide, protein or other macromolecule.)

[ValWood]

oh right GO:0043041 amino acid activation for nonribosomal peptide biosynthetic process

it probably isn't a particularly meaningful process grouping term? and it isn't used in annotation, so obsoletion might be better?

[hdrabkin]

unless it would be needed to house the non-tRNA dependent aa activations -(

[ValWood]

won't GO:0043041 amino acid activation for nonribosomal peptide biosynthetic process do that?

[hdrabkin]

Is that a child or aminoacid activation or of aminoacyl-tRNA activation?

[hdrabkin]

That is; amino acid activation for nonribosomal peptide biosynthetic process can include aa-tRNA activation also (eg, aa-TRNA protein transfer still results in a peptide bond...

[ValWood]

These seem like common activities, that should be connected at the MF level, but not at the process level.

.... non-ribosomal peptide synthesis is a totally different process to translation (genes, species, products etc, etc, etc....... so this seems a strange axis of classification in the process ontology? it's not really a biologically related "module" other than both processes produce a peptide bond......this seems like an over-complication in Biological Process....

@pgaudet ?

[hdrabkin]

The whole mess started because people think that tRNA aminoacylation should have some connection to translation, so the terms aa-tRNA activation for translation was made, and then anything. Once you get the aa-tRNA, one can use it for translation; or other things, depending upon which aa-tRNA you have. Yes it's complicated because aa-TRNAs have been hijacked several times. The problem term is the 'amino acid activation', because it turns out that there is more than one way to activate an amino acid.

[ValWood]

Hi Harold.

This can be fixed easily.

1. We can remove "amino acid activation" (obsolete). It isn't required. It isn't used in direct annotation. Everything is annotated to one of the children. It isn't necessary to couple "activation" for translation and non-ribosomal peptide synthesis? The important thing is that tRNA synthases involved in translation are coupled to translation (for enrichment and slimming). Everything will be fine if you remove this term, it serves no useful purpose for biological process grouping. This will remove the problem connection to amino acid metabolism.

2. Should these even be a term "tRNA aminoacylation for nonribosomal peptide biosynthetic process" this process doesn't use tRNAs does it? (it has no annotations!)

There are probably too many terms around here that could be handled in other ways now that we can couple MF to BP easily, but this will resolve the immediate problem of parentage causing these gene products to be annotated to "amino acid metabolism"

Would that work for you? Or am I missing something?

[ValWood]

Even the def of this term is a problem GO:0043038 amino acid activation The modification of an amino acid to an active form, for incorporation into a peptide, protein or other macromolecule.

it seems to fit "activation for non ribosomal peptide synthesis" better than "activation for translation" (I would not call tRNA charging for peptide synthesis and amino acid modification?

These changes would make GO:0043041 amino acid activation for nonribosomal peptide biosynthetic process a bit redundant too.

You would only need GO:0043042 amino acid adenylylation by nonribosomal peptide synthase unless there is another way to activate amino acids for nonribosomal peptide biosynthetic process?

"activation" seems a bit "wooly" we are forcing a connection between processes that are not really biologically related at the process level. They have more in common at the level of molecular function.

[ValWood]

PS I realise that aminoacylated tRNAs are used in other contexts, but the alternative processes aren't represented in the graph here as far as I can see? I know I have an example of this but I can't find it....

[hdrabkin]

It appears that Cyclodipeptide synthases use an amino-acyl tRNA as a substrate. However, there are two ways to synthesize a cylodipeptide, depending organisms

Below is summary from PMID: 21325056

1. Most cyclodipeptide biosynthesis pathways involve nonribosomal peptide synthetases (NRPSs) ( NRPSs are large multifunctional modular enzymes in which each module catalyzes a single step of peptide synthesis by incorporating one amino acid. In the standard NRPS module, the peptide bond-forming condensation domain (C) and the amino acid adenylation domain (A) are linked through a phosphopantetheine-tethered domain (T) to synthesize the peptidyl chain as a growing thioester. The enzyme usually ends with a thioesterase domain (TE), which releases the peptidyl chain by hydrolysis or cyclization (5). (this is more in line with GO:0043042
2. However, in 2002 Lautru et al. identified the albC gene in the genome of the bacterium Streptomyces noursei, and its product AlbC, which lacked similarity to the NRPSs, was able to catalyze the formation of a cyclic dipeptide (6). AlbC is the first enzyme in the biosynthetic pathway of albonoursin, a secreted antibacterial DKP bearing a cyclodipeptide structure. Recently, AlbC was shown to use aminoacylated tRNAs as substrates instead of free amino acids (7), thus allowing the enzyme to skip the amino acid activation step usually performed by the adenylation domain in NRPSs. This is more in line with GO:0043040.

So, aa-tRNA is used by one, but not all paths to a cyclic dipeptide. And thus, there are TWO ways to activate an amino acid for non-ribosomal peptide synthesis, but only one way for ribosomal peptide synthesis.

So if I obsolete GO:0043038 (amino acid activation) , the link to amino acid metabolism is gone, but then GO:0043041 has no is_a parent. (has a part_of to GO:0019184). GO:0043042 is_a GO:0043041 as is GO:0043040 I need an is_a for GO:0043041 ??

[ValWood]

bloody biology!

so what is "activation" molecularly? "The modification of an amino acid to an active form, for incorporation into a peptide, protein or other macromolecule." what does the modified part mean? This might help to decide...

[deustp01]

"Activation" of a molecule means something like converting the molecule to a more chemically reactive form. Phosphate is activated when it's the terminal phosphate group of ATP. An amino acid can be activated by conjugating it with coenzyme A, as can a fatty acid. The activated form then can participate in reactions in which the original molecule (phosphate, amino acid, fatty acid) is covalently attached to something else in a reaction that would otherwise be energetically very unfavorable, (Fritz Lipmann and colleagues worked out the structures of ATP and coenzyme A as part of trying to understand how phosphorylation and fatty acylation work, and invented the term "high energy bond" to describe the linkage formed in the activation reaction that allowed the subsequent conjugation reaction to proceed.)

[hdrabkin]

yes, so aa-AMP and aa-CoA are 'activated forms of the amino acid. For tRNA, needed for the transfer or the aa to the terminal adenosine moeity of the tRNA. I

[ValWood]

They aren't really "processes" are they?

Perhaps get rid of BOTH GO:0043038 AND GO:0043041

GO:0043042 the only thing without an is_a is "amino acid adenylylation by nonribosomal peptide synthase"

this one could be is_a amino acid metabolism? and part_of "nonribosomal peptide biosynthetic process"

[ValWood]

Should I close this ticket so in future we don't need to read all the 35! comments.

I suggest a new ticket to propose obsoletion of GO:0043038 amino acid activation GO:0043039    tRNA aminoacylation GO:0043041    amino acid activation for nonribosomal peptide biosynthetic process

because it represents a Molecular function in the biological process ontology

@hdrabkin @pgaudet @deustp01 do you agree?