gocentral
commented
21 years ago Logged In: YES user_id=436423
- The issue of grouping terms was discussed at the January 2003 meeting, so I'll leave it to be addressed elsewhere (there should be more SF entries for individual cases as we deal with them).
- The parent-child relationship between GO:0019839 and GO:0003747 looks like an error, because 'translation release factor activity' refers to releasing a polypeptide chain, not a guanyl nucleotide. I'll remove the relationship.
- As far as I can tell, 'guanyl-nucleotide release factor' is the same as 'guanyl-nucleotide release factor,' GO:0019839 should be merged into GO:0005085. This will also take care of the last point, because guanyl nucleotide binding activity (GO:0019001) is already a parent of guanyl-nucleotide exchange factor activity (GO:0005085).
m
Original comment by: mah11
I'm trying to annotate a protein, MOG1, that binds to the yeast ortholog of RAN, a GTPase, and the term that seems the most appropriate specific description of its molecular function is 'guanyl-nucleotide release factor' GO:19839 (not defined).
However, at present, using this would create a true path violation because one of the parents of 19839 is 'translation release factor'. It seems to me that this is the wrong way around, that a 'translation release factor' that involves stimulating release of a guanyl-nucleotide should be an instance of a more general 'guanyl-nucleotide release factor'.
In addition, the other main parentage branch of 19839 is from 'signal tranducer' (GO:0004871) and its child 'receptor signaling protein' (GO:0005057). This parentage also seems perhaps inappropriate, as the 'guanyl-nucleotide release factor' that I want to annotate is involved in the nucleus to cytoplasm cycling of a protein that is involved in nuclear import. Current paths to root are at the bottom of this mail
Is 'signal transducer' GO:0004871 (an old term from the original set) really a molecular function anyway? It seems to me that this might really be a process, since it intrinsically involves the interactions of multiple proteins.
If 'guanyl-nucleotide release factors' are involved in so many processes, I think we this term should not be a child term of these "process-like" terms like 'translation regulator' and 'signal transducer'
On a bigger picture type view, I think this points out a danger of certain grouping terms in the function ontology whose purpose is to keep things involved in the same process together. This is what worries me about keeping the grouping term for complement stuff, to refer to a previous discussion. They can destroy the orthogonality of the function and process ontologies. This is a really good example of a case were one of the first studied uses of a function, GDP/GTP exchange factors, in a given process, i.e. signal transduction, is not the only process that the function is used in.
In any case, it looks at the moment like all of the guanyl-nucleotide
exchange factor, release factor type terms are in this same branch under
'signal transducer ; GO:0004871'. Isn't GTP/GDP exchange used in vesicular transport too?
Does anyone know if 'guanyl nucleotide binding ; GO:' is an appropriate parent for 'guanyl nucleotide release factor'? It isn't a parent currently. A current parent of the release factor term is 'small GTPase regulatory/interacting protein'; should it also have 'guanyl nucleotide binding' as a parent? In any case, I don't think this term should be under 'signal transduction either'. I don't really have a good proposal for appropriate changes yet, I just wanted to put this out there to get input and start the discussion. I'd really like to annotate MOG1 to the function term 'guanyl nucleotide release factor' but feel I can't use it with its current parentage.
Thanks,
-Karen
$Gene_Ontology ; GO:0003673
<molecular_function ; GO:0003674 %binding ; GO:0005488 ; synonym:ligand %nucleic acid binding ; GO:0003676 %translation factor\, nucleic acid binding ; GO:0008135 % translation reg %translation termination factor ; GO:0008079 %translation release factor ; GO:0003747, GO:0003748, GO:0003749 %guanyl-nucleotide release factor ; GO:0019839 % small GTPase regul
%signal transducer ; GO:0004871 %receptor signaling protein ; GO:0005057 %small GTPase regulatory/interacting protein ; GO:0005083 %guanyl-nucleotide release factor ; GO:0019839 % translation release f
%translation regulator ; GO:0045182 %translation factor\, nucleic acid binding ; GO:0008135 % nucleic acid bind %translation termination factor ; GO:0008079 %translation release factor ; GO:0003747, GO:0003748, GO:0003749
%guanyl-nucleotide release factor ; GO:0019839 % small GTPase regulat
Reported by: krchristie
Original Ticket: "geneontology/ontology-requests/560":https://sourceforge.net/p/geneontology/ontology-requests/560