search

geneontology / go-ontology

Source ontology files for the Gene Ontology
http://geneontology.org/page/download-ontology
Creative Commons Attribution 4.0 International
223 stars 40 forks source link

revise LINC complex definition #20198

Open ValWood opened 4 years ago

ValWood commented 4 years ago

Sad1-Kms1 LINC complex The current definition is bad, and the parentage is messed up.....my fault. Existing def A LINC complex implicated in the connection of DNA double strand breaks to the cytoskeleton during DNA double-strand break repair. PMID:24943839 PMID:24947240

PMID:24943839 PMID:24947240 PMID:27889481

remove parent DNA repair complex (indirect) remove connection to DNA repair

ValWood commented 4 years ago

GO:0106094 nuclear membrane microtubule tethering complex This is describing LINC complexes Revise def Any of the conserved LINC (linker of nucleoskeleton and cytoskeleton) complexes which connect the nuclear outer and inner membranes together, and link the nuclear lumen to some part of the cytoskeleton (e.g. MTOC, microtubules or actin). LINC complexes comprises of a KASH-domain outer NE protein and a SUN-domain inner NE protein, and may include additional accessory proteins. PMID:1922512 PMID:24943839 PMID:24947240 PMID:27889481 relabel LINC complex make nuclear membrane microtubule tethering complex narrow synonym

Definition (GO:0034993 GONUTS page) Revise def Any of the conserved LINC (linker of nucleoskeleton and cytoskeleton) complexes which connects the nuclear outer and inner membranes together, and connects the centrosome with telomere bouquet during meiotic prophase. exact synonym Sad1-Kms1 LINC complex

pgaudet commented 4 years ago

@ValWood

I did the first 3 actions in the previous comment. I can create NEW mitotic centromere-centrosome LINC complex - but I wonder if you want changes to the sibling 'GO:0106084 mitotic nuclear membrane microtubule tethering complex' ?

Or should I change GO:0106084 instead of creating a new term ?

ValWood commented 4 years ago

I would ignore the term GO:0106084 for the time being- I don't really know what it refers to. It's SPC29, which is a component of the inner plaque of the SPB in S. cerevisiae for which we have not detected an ortholog in other species, plus another non-conserved protein (Mps2) one component of NMS complex (SOS7). These are all sub parts of the SPB and it is possible that they facilitate the connection to the LINC complexes in some way (ie they might be the SPB components that the LINC complex adapts to). The composition of the complex is odd because these proteins are spatially separated (i.e inner plaque and outer plaque).

Either way it isn't a LINC complex which is evolutionarily conserved and has a defined composition.

ValWood commented 4 years ago

even more confusing GO:0106084 mitotic nuclear membrane microtubule tethering complex Mps2-Bbp1-Spc29 complex and Comments In S. cerevisae Mps2-Bpb1 is the membrane anchor sub complex and Spc29 is the spindle pole body linker molecule.

BUT mps2 in budding yeast is not annotated to this complex. It is annotated to meiotic nuclear membrane microtubule tethering complex (by IMP which is also odd?)

Maybe, if SGD agrees this one could be obsoleted as it's meaning is very unclear?

ValWood commented 4 years ago

Actually, it has no annotation, so it's clearly unclear what it is for... It seems that Intact probably requested it originally looking at the history log so maybe @bmeldal can confirm?

bmeldal commented 4 years ago

I've sent an email to @sandraorchard as she curated this entry.

pgaudet commented 3 years ago

@sandraorchard did you have a chance to look at this ?

bmeldal commented 3 years ago

PMID:28356353

Ok, looks like Mps2-Bpb1 forms an anchor between the nuclear envelope (NE) and the central plaque of SPB, by way of Bbp1-Spc29 interaction to anchor to the central plaque, and interaction with NE components Mps3 and Ndc1.

Sandra picked apart the SPB into smaller subunits and therefore curated the anchor as a separate complex.

Whether it should be a complex in it's own right is debatable. One could argue that the whole SPB should be a complex/GO term and the three plaques shouldn't be separate terms either ;-)

There are no annotations as we link the complex to the GO term but that's an equivalence statement that doesn't go into the annotations. We'd have to ask SGD to make the relevant GP annotations.

pgaudet commented 3 years ago

@bmeldal

What do you mean, Sandra annotated to subcomplexes ? I see CPX-1287 | mps2-bbp1_yeast annotated to GO:0005816 | spindle pole body from PMID:28356353

In any case, it looks like you didn't need GO:0106084 mitotic nuclear membrane microtubule tethering complex ?

Thanks, Pascale

pgaudet commented 3 years ago

How about those as action points:

Is that right ?

bmeldal commented 3 years ago

MPS2-BBP1 spindle pole body anchor complex (https://www.ebi.ac.uk/complexportal/complex/CPX-1287), defined as "Key mediator of the link between the nuclear envelope and the spindle pole body (SPB), anchoring the SPB to fusion sites of the inner nuclear membrane and outer nuclear membrane. Forms higher molecular weight assemblies with other SPB proteins (MPS3/P47069, SPC29/P33419) which hold the SPB in the nuclear envelope."

has been annotated as GO:0106084 mitotic nuclear membrane microtubule tethering complex.

But it doesn't show up as a GO annotation as the relationship is CPX-1287 = GO:0106084, not CPX-1287 part_of GO:0106084. These equivalence statements are excluded from the GO annotations but exist in CP.

You find the link toward the bottom of the page under "cellular location".

@ValWood would MPS2-BBP1 qualify for your LINC complex term?

I'm no expert in the SPB. These large machines give me headaches as it's hard to distinguish what's a complex within the whole machine!

ValWood commented 3 years ago

@bmeldal MPS2-BBP1 isn't a LINC complex becasue a LINC complex has SUN and KASH domain proteins. https://en.wikipedia.org/wiki/LINC_complex when it is linked to MSP3 (SUN domain)

I'm not sure what MPS2-BBP1 is because it does not appear to be well conserved. but https://pubmed.ncbi.nlm.nih.gov/28356353/, the most recent paper

Reconstitution of the Mps3-Mps2-Bbp1 complex "......Thus, Mps3 forms a stable complex with Mps2-Bbp1".

so maybe MPS2-BBP1 is only part of the complex?

ValWood commented 3 years ago

Review https://www.mdpi.com/2073-4409/8/10/1285

LINC complexes

Screenshot 2021-02-10 at 18 14 52

linC complexes span the inner and outer membrane.

and that is how "nuclear membrane microtubule tethering complex" is defined A nuclear membrane protein complex which connects the nuclear outer and inner membranes together, and links thereby links the nuclear lumen to cytoplasmic microtubules. [PMID:19225124]

I'm not aware of any other complexes, but to accommodate "LINC complex" should be a child of this term.

I would do:

GO:0106094 nuclear membrane microtubule tethering complex
--NEW LINC complex (As above but with SUN and KASH) ----GO:1990612 Sad1-Kms1 LINC complex ----GO:0034993 meiotic nuclear membrane microtubule tethering complex -> rename -> meiotic LINC complex

remove the parents DNA repair complex and GO:0034993 meiotic nuclear membrane microtubule tethering complex from GO:1990612

pgaudet commented 3 years ago

ah ! but you had asked to merge Sad1-Kms1 LINC complex - seem comment above https://github.com/geneontology/go-ontology/issues/20198#issuecomment-717861981

You dont want that anymore ?

ValWood commented 3 years ago

Gosh, I'm not sure. Whatever we do here won't work very well. The existing def of GO:1990612 Sad1-Kms1 LINC complex is too narrow. Maybe it is better not to have the specific complexes and only have a generic LINC complex. I'm also not sure about the conservation, I was trying to make it generic but that is difficult too because the conservation is poor.

Can we wait on this one. I still never completed my review of nuclear envelope tethering and that will need to wait until after grant writing.

pgaudet commented 3 years ago

sounds good. I reassigned to you, and added the 'waiting for feedback' label. You can remove that and add 'ready whenever that's the case.

Thanks, Pascale