3.6.4.12 not DNA helicases

[ValWood]

P87114 SPAC20G8.08c SMARCAD1 family ATPase Fft1

GO:0003678 | DNA helicase activity | IEA with 3.6.4.12

@pgaudet @colinlog

[ValWood]

ACtually @pgaudet here it is the EC assignment to the protein that is correct, not the mapping. WHo does these?

[ValWood]

same situation for SPAC11E3.01c SNF2 family ATP-dependent chromatin remodeller Swr1 O13682

[pgaudet]

I assign @sylvainpoux these tickets. @sylvainpoux please let us know if you'd like to assign other people.

Thanks, Pascale

[sylvainpoux]

Hi Val,

I'm sorry, but I don't understand your message (it is rather cryptic). May I ask you to detail the issue? Are there publications? Is it the EC to GO mapping that causes a problem?

Thanks

Sylvain

[ValWood]

I think it is the EC assignment rather then the mapping. P87114 and O13682 are assigned the EC number for DNA helicase, but they aren't DNA helicases.

(they have the a domain that is called 'helicase' but they are not helicases in vitro.

SMARCAD1 family are known to be ATP-dependent chromatin remodeler activity (GO:0140658) but their role is not to unwind DNA.

[sylvainpoux]

Hi Val,

Thanks for your explanation. We will look more in detail. May I ask you to send a user request to UniProt? A curator will look to these proteins.

By the way, ATP-dependent chromatin remodeling activity can imply DNA helicase activity, since DNA unwinding is usually required for chromatin remodeling.

Thanks

Sylvain

[ValWood]

I will mail the helpdesk.

ATP-dependent chromatin remodeler activity is a new term, defined

Definition (GO:0140658 GONUTS page) An activity, driven by ATP hydrolysis, that modulates the contacts between histones and DNA, resulting in a change in chromosome architecture within the nucleosomal array, leading to chromatin remodeling. PMID:14729263 PMID:21862382 PMID:30867599

It is meant to exclude 'helicase activity' will be a sibling. These chromatin remodelers are no longer considered to be helicases in vitro. Apparently, the energy here is not to unwind DNA directly. IN think Ino80 stabilized rerplication forks in some way, but doesn't unwind DNA?

@colinlog can probably summarize betterr.

[ValWood]

I should also explain that ATP-dependent chromatin remodeler activity is for the activity.

These would still be annotated to the "chromatin remodelling" process. But the helicase family proteins involved in chromatin remodelling do not appear to be helicases.

[sylvainpoux]

ok I understand now.

Interesting: same ATPase reaction (Rhea) but different EC number.

We always come back to the same problem of EC numbers that correspond to more than one different reaction (protein kinases) or one reaction with many EC numbers (this case)

Thanks

Sylvain

[pgaudet]

RHEA ATPase should now be mapped to ATP hydrolysis activity.

[ValWood]

OK so it is a mapping issue after all.

[pgaudet]

I am not sure I follow:

P87114 has the following keywords:

• Chromatin regulator,
• DNA-binding,
• Helicase,
• Hydrolase Helicase (KW-0347) is mapped to GO helicase activity [ GO:0004386 ]

Likewise, O13682 also has Helicase as a keyword.

It seems that the KW2GO mapping is correct, and the keyword assignment is leading to these annotations.

Isn't it, @sylvainpoux ?

Thanks, Pascale

[sylvainpoux]

Hi Pascale, yes, it is a problem of keyword. We annotated the ATPase activity with the EC number for helicase activity. However, it is apparently not an helicase, but an ATP-dependent chromatin remodeling protein. We will update these entries and keep the ATPase activity without adding the EC number for helicases. Thanks Sylvain

[pgaudet]

Thanks! From what I see in QuickGO the KW also needs to be removed.

[sylvainpoux]

Hi Pascale, yes, the KW should be removed as well S.

[cmungall]

Any update on this? It looks there there are still two errors:

• P87114 to EC:3.6.4.12
• P87114 to KW:Helicase

I think the issues also extend to orthologs, which are also ATP-dependent chromatin remodeler, e.g. YEAST FUN30

https://www.uniprot.org/uniprotkb/P31380/entry

And human SMARCAD1

https://www.uniprot.org/uniprotkb/Q9H4L7/entry

See this paper: https://amigo.geneontology.org/amigo/reference/PMID:22960744

(as an aside, I think more confirmatory yeast annotations could be made from this paper, e.g. The ATPase activity of Fun30 is essential for its chromatin remodelling activity. Expression of wild-type Fun30, but not ATPase-dead Fun30K603R in fun30Δ restored end resection to wild-type levels (Fig 2c). This suggests that chromatin remodelling driven by Fun30 facilitates long-range resection, either directly or indirectly., suggests chromatin remodeling ATPase activity, evidence against helicase)

[ValWood]

It probably didn't get assigned to anybody. I can't trace the e-mail I said I would send to the UniProt helpdesk. Tagging @Antonialock