Closed sjm41 closed 3 years ago
@hdrabkin Appreciate your thoughts on this one too.
So the EC appears to just be mechanistic; the other GO terms are substrate specific. The EC parent of 3.6.1 GO:0003924 GTPase activity (RHEA 19669, which maps back to) ARGH! 19669 is the GTP specific substrate rexn; The EC 3.6.1 is only found in an obsoleted GO term which was a grouping term specific to GTP.
@kaxelsen Can you provide any extra insights on the above? Is https://enzyme.expasy.org/EC/3.6.1.15 meant to be for a specific enzyme/reaction (which is how the EC comments read), or more generic/mechanistic (which is how the EC name and reaction read, and is how the xreffed GO term is used). Thanks.
All EC numbers describe specific proteins. That means that if an EC number is associated with a generic reaction, it is because the enzyme it describes has a broad substrate specificity. Likewise, an EC number associated with a very specific reaction describes an enzyme with very narrow substrate specificity (or that will only see that one substrate in vivo). An EC number will never be a parent for another EC number, e.g. EC 1.1.1.1 associated with the reaction "an alcohol + NAD+ = an aldehyde + NADH + H+" is not the parent for all the EC numbers describing the oxidation of specific alcohols with NAD+, it is just an alcohol dehydrogenase that oxidises all alcohols.
So I think you should remove the EC number from GO:0017111, but you can keep the reaction
in that case the RHEA should be copied to the def ref as the reference, and the EC removed from def and xref
Thanks @kaxelsen - makes sense! It did look odd to see this GO term with a 4-digit EC xref having several GO children with other 4-digit EC xrefs. As you both say, removing EC:3.6.1.15 from GO:0017111 will resolve and clarify things.
@kaxelsen - one other thing, looking the EC parent here (and how it relates to GO structure):
EC:3.6.1.- is "Hydrolases -> Acting on acid anhydrides -> In phosphorus-containing anhydrides" but there are also these 'sister' groupings: EC:3.6.4.- Hydrolases -> Acting on acid anhydrides -> Acting on ATP; involved in cellular and subcellular movement EC:3.6.5.-Hydrolases -> Acting on acid anhydrides -> Acting on GTP; involved in cellular and subcellular movement
Aren't ATP and GTP specific examples of "phosphorus-containing anhydrides"? Naively, that makes me think those ECs would be subclasses of 3.6.1.-? But maybe it's the "involved in cellular and subcellular movement" aspect that distinguishes these classes and makes them a separate 3-digit grouping? Do you mind clarifying that for me, thanks!
GO:0017111 fixed; will close ticket; Other issue maybe goes in separate ticket?
Thanks Harold. Yes, I'll pursue my wider query about EC:3.6.-.- with Kristian elsewhere (if he doesn't answer here!) and make a new ticket if required.
@sjm41 It is exactly as you say. The enzymes in EC 3.6.4- and EC 3.6.5.- are all involved in movement. Note that most of the EC numbers earlier classified here have been transferred to the new EC 5.6.-.- (Isomerases altering macromolecular conformation) since the movement involves conformational changes.
Upon review, we added EC:3.6.1.15 back as an xref to 'nucleoside-triphosphatase activity' (GO:0017111) in #23401. EC:3.6.1.15 exactly matches the GO definition and the RHEA xref (RHEA:23680). Moreover, there are lots of other examples in the GO where we have a 4-digit EC (corresponding to an enzyme with broad specificity) with children that also have 4-digit ECs (corresponding to enzyme with restricted specificity).
Currently there is:
id: GO:0017111 name: nucleoside-triphosphatase activity namespace: molecular_function def: "Catalysis of the reaction: a nucleoside triphosphate + H2O = nucleoside diphosphate + phosphate." [EC:3.6.1.15] xref: EC:3.6.1.15 xref: RHEA:23680 is_a: GO:0016462 ! pyrophosphatase activity
Problem is that EC:3.6.1.15 seems to be for a specific enzyme (https://enzyme.expasy.org/EC/3.6.1.15), while the GO term is used as a grouping term for 6 is_a children with different EC/RHEA xrefs: GO:0050333 thiamin-triphosphatase activity (EC 3.6.1.28 / RHEA 11744) GO:0003774 motor activity GO:0140603 ATP hydrolysis activity (RHEA 13065) GO:0043273 CTPase activity GO:0003924 GTPase activity (RHEA 19669, which maps back to EC:3.6.5.-) GO:0050339 thymidine-triphosphatase activity (EC 3.6.1.39 / RHEA 19013)
So, is "GO:0017111 nucleoside-triphosphatase activity" meant to be broader than EC:3.6.1.15? If so, then we should remove its EC:3.6.1.15 xref and create a new, separate term for EC:3.6.1.15 under GO:0017111.
Related point - if GO:0003924 GTPase activity accurately maps back to EC:3.6.5.-, then maybe GO:0003924 needs moving to be a direct child of child of "hydrolase activity, acting on acid anhydrides (GO:0016817)" (= EC:3.6)??