geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
216 stars 39 forks source link

Preribosome query #16684

Open pgaudet opened 5 years ago

pgaudet commented 5 years ago

From the discussion in #16628 (preribosome assembly )

I wanted to ask @bmeldal if you consider GO:0030684 preribosome a real complex ? There are 4 EXP.

There are several children with EXP annotations: 150 EXP preribosome, large subunit precursor 97 EXP small-subunit processome 95 EXP 90S preribosome 18 EXP preribosome, small subunit precursor 17 EXP PeBoW complex 12 EXP Pwp2p-containing subcomplex of 90S preribosome 6 EXP UTP-C complex 5 EXP Mpp10 complex 4 EXP preribosome 3 EXP Noc1p-Noc2p complex 2 EXP Noc4p-Nop14p complex 2 EXP nucleolar preribosome 1 EXP Noc2p-Noc3p complex

What do you think of these terms ?

Thanks, Pascale

bmeldal commented 5 years ago

Well, it's a bit dodgy and looking at the log it's old. I can see why it was created in 2003, to group all pre-assembly ribosomal complexes. However, since there are only 4 exp something tells me people were not very keen on using it ;-) I'd re-annotate to GO:1990904 ribonucleoprotein complex and the appropriate child of GO:0042255 ribosome assembly.

As for the children, some look a but dodgy, too (large/small subunit precursor, nucleolar pre-ribosome). I'd say, as a rule of thumb, if it's a bona fide, and well-defined, pre-assembly complex they should be made if requested as they all have some specific function. They are a bit like the splicosome or the TFs, the composition changes during the assembly but they have specific functions (such as GTPases...) to get to the fully functional ribosome. And many proteins only occur in the pre-assembly complexes. I have worked on the cap-binding complexes already and the translation initiation complexes are "next" so I'll be coming back to this for sure!

pgaudet commented 5 years ago

Thanks @bmeldal So, for now, should we obsolete 'GO:0030684 preribosome' with 'consider GO:1990904 ribonucleoprotein complex'?

And move all these children to GO:1990904 ribonucleoprotein complex '?

Pascale

bmeldal commented 5 years ago

So, for now, should we obsolete 'GO:0030684 preribosome' with 'consider GO:1990904 ribonucleoprotein complex'?

Yes

And move all these children to GO:1990904 ribonucleoprotein complex '?

At first glance only the 3 mentioned above (large/small subunit precursor, nucleolar pre-ribosome). I would have to look more closely at the other children if they are all bona fide assembly complexes.

mah11 commented 5 years ago

On a related note, do any of you know whether the "Nsa1 particle" would meet current criteria for a GO CC term? I've just heard about it curating PMID:27667686, which cites PMID:21763358. I'm not sure how clearly it could be defined, as it looks from PMID:21763358 figure 1 as though it's a 60S precursor with some extra proteins stuck on (which dissociate before the 60S subunit is mature).

p.s. I'm a little confused about whether the large and small subunit precursor terms (GO:0030687 and GO:0030688) are to be moved or obsoleted. If they remain, I'll use GO:0030687 in a phenotype logical definition.

bmeldal commented 5 years ago

do any of you know whether the "Nsa1 particle" would meet current criteria for a GO CC term?

If it has a defined role during ribosome maturation, then I think it's a bona fide term. But I haven't looked into the literature for it yet.

p.s. I'm a little confused about whether the large and small subunit precursor terms (GO:0030687 and GO:0030688) are to be moved or obsoleted.

I regard these terms as bucket terms for a list of specific ribosomal precursor complexes. They probably don't share the same function, hence the suggestion to obsolete them and where possible, use the specific precursor complex terms.

NB: We haven't tackled them yet for the CP except for the yeast large and small subunit ribosomal units. But in Reactome terms, they would all be good complexes.

mah11 commented 5 years ago

Hmm, I don't know this area well at all; I've just blundered into it via reviewing community curation for PMID:27667686. That paper draws a model (below) in which "Nsa1 particle" and "Rix1 particle" (not to be confused with GO:0097344 ! Rix1 complex, as far as I can tell) are distinguishable intermediates in large subunit assembly. But I don't have the background to tell how much of that is speculation vs. reasonable interpretation of their data vs. already established knowledge. I also can't tell whether these intermediate particles have any role other than "step on the way to becoming a functional ribosome".

At the moment we would probably accept the small loss of an annotation or two, rather than take the time to dive down the rabbit hole. It simply didn't hurt to check whether someone already knew enough to sort it out!

gr7

pgaudet commented 5 years ago

Maybe as a first step we can obsolete 'preribosome' and 'nucleolar preribosome', since these have few annotations are are unlikely to represent a specific complex. What do you think ?

ValWood commented 1 year ago

ACtion Maybe as a first step we can obsolete 'preribosome' and 'nucleolar preribosome', since these have few annotations are are unlikely to represent a specific complex.