chromocenter (GO:0010369)

[ValWood]

A region in which centric, heterochromatic portions of one or more chromosomes form a compact structure.

I never came across this term in a paper. How does it differ from pericentric heterochromatin (GO:0005721)

Heterochromatin that is located adjacent to the CENP-A rich centromere 'central core' and characterized by the modified histone H3K9me3.

[ValWood]

I ask because I am getting PAINT annotation. I will filter this, and only request an annotation fix if the term remains

[ValWood]

chromocenter 68 EXP pericentric heterochromatin 104 EXP

If they are the same I prefer the primary term name 'pericentric heterochromatin' because this is universal

[mah11]

I don't think they're identical.

'chromocenter' has 479 hits in PubMed

From a quick skim, I think the distinction is that a chromocenter is a bunch of constitutively heterochromatic regions (some abstracts seem to differ on whether it can include non-centromeric hc, but they seem to consistently say that it can contain more than one centromere), piled together within the nucleus. They're found in various species including plants, flies, and mammals, but the term may well be irrelevant to fungi, or at least fission yeast.

sample PMIDs: 29458329, 29626931, 32182352, 32393416, 32161227

[ValWood]

The centromeres from different chromosomes do cluster together but I have never heard anyone use a 'component' term to describe this concept in fission yeast. Since all centromere components would be a part of the chromocenter (if a chromocenter is all of the centromeres when they are connected) a single annotation looks odd.

Q Is this really a bona fida cellular component? or a component epithet for a function/process (i.e. GO:0098653 centromere clustering)

Even if a "chromocentre" can be described semantically, is it useful annotation object?

This term bothers me because it will cause lot of logic issues related to chromosome segregation. I can elaborate later if necessary.

[ValWood]

OK here is the first logical issue:

A chromocenter is not part of a single chromosome. I'll leave it there for now...

[ValWood]

Adding obsoletion label. I hope nobody objects...

[ValWood]

abstract of the first paper in @mah11 list above

Chromocenters are defined as a punctate condensed blocks of chromatin in the interphase cell nuclei of certain cell types with unknown biological significance. In recent years a progress in revealing of chromocenters protein content has been made although the details of DNA content within constitutive heterochromatin still remain unclear. It is known that these regions are enriched in tandem repeats (TR) and transposable elements. Quick improvement of genome sequencing does not help to assemble the heterochromatic regions due to lack of appropriate bioinformatics techniques.

This definition indicates that "chromocenter" is slightly orthogonal to a GO component.

[hdrabkin]

So the children will need to go if there is no such thing?

[ValWood]

Oh. It has kids !!!!???

[ValWood]

For one GO:0005701 polytene chromosome chromocenter A region at which the centric regions of polytene chromosomes are joined together.

I'm pretty sure there is an existing consensus to get rid of polytene chromosome. Checking with @hattrill @pgaudet Did I remember this correctly? If not, is this a good idea? Now I am confused if this is an inter or intra chromosome term

@hdrabkin I did not look at the other child yet...

[hattrill]

This article https://academic.oup.com/nar/article/48/8/4161/5809156 was written in 2020, and describes chromocentres and does not appear to be synonymous with pericentric heterochromatin.

"Molecularly, chromocenters are composed of pericentromeric heterochromatin and CenpA containing centromeric chromatin, which are both rich in repetitive DNA and evolutionarily highly dynamic "

As this structure is actively being studied and seems to have biological relevence, I would like to keep it. If the structure of the ontology is a problem, perhaps that is where the fix should lie.

(as for the polytene issue - on my radar, but it's big sized project for me to do at some other time - more pervassive than just GO annotation.)

[hdrabkin]

From what little perusing I was able to do at Pubmed, I too am reluctant to have it obsoleted. Perhaps if yeast/fungi do not have them, might want to just add a taxa restriction? However, I see your point about not having a suggestion in the ontology that they are part of a single chromosome.

[ValWood]

OK good to move it because it isn't part of a single chromosome.

I'm not saying that yeast don't have them (they do) but I will probably avoid using it because to me it doesn't totally make sense as a component...."chromocentres" are chromosome organizations, This would be similar to a term for say "Rabl configuration" in Interphase, or for the "telomere bouquet". We do not have a cellular componet term for these "multi-chromosome arrangement" although we have process terms to describe their formation.

i.e. centromere clustering at the mitotic interphase nuclear envelope http://amigo.geneontology.org/amigo/term/GO:0072766

(I'm also struggling to imagine how you would usefully annotate a gene product to a 'chromocentre')

If kept it needs to be re-housed somewhere near the root node.

[hdrabkin]

Any suggestions as to where to move it would be appreciated. @hattrill, any ideas?

[ValWood]

You can only put it under "cellular componet"

[ValWood]

Well you could put it under intracellular.

[hdrabkin]

would it be fair to put it under the nucleus?

[ValWood]

Not if it is present during mitosis

[hattrill]

Seems to fit with the bunch of misfits under: GO:0043232 | intracellular non-membrane-bounded organelle

[hattrill]

(I'm also struggling to imagine how you would usefully annotate a gene product to a 'chromocentre')

If I curated this paper....https://academic.oup.com/nar/article/48/8/4161/5809156

[hdrabkin]

Ah, @hattrill Iike that idea.

[ValWood]

OK makes sense if it's isolated.

[ValWood]

Definition (GO:0010369 GONUTS page) A region in which centric, heterochromatic portions of one or more chromosomes form a compact structure. PMID:12384572 PMID:15053486 PMID:16831888

Is this definition correct? Can a single chromosome really be a chromocentre?

[hattrill]

My understanding is that they are formed by the bundling together of chromosomes.

[ValWood]

So the current definition is incorrect?

It seems that it would make sense to define as more than one chromosome kinetochore? That way we can justify it not being housed under a "chromosomal region" which would cause problems for the common usage of more than one kinetochore piled together.

Also, I don't understand how a kinetochore on its own (or a pair of kinetochores) from a single chromosome could be a "chromocentre" . Thats just a duplicated pre-segregated chromosome ?

So actions, if in agreement

1. Fix definition

A region in which centric, heterochromatic portions from more than one chromosomes form a compact structure.

2. Move to GO:0043232 | intracellular non-membrane-bounded organelle

Not that: Although the structure technically exists, because the centromers are clustered at specific points in the cell cycle, I'm not going to use it for fission yeast. It would be a headache because to annotate consistently all of the 100's of constitutive kinetochore components would need to be co-annotated to "chromocentre", and this would not add anything for our users. Anything which is part of the kinetochore at specific point in the cell cycle is necessarily part of a cluster/chromocentre. I don't think of this as a "cellular component" but a construct to enable researchers to talks about "all of the kinetochores clustered together in interphase"

By analogy, we could aslo invent a similar componet term to talk about all of the chromosomes at the same time during clustering "genome center?".

I would rather represent the 'process' of clustering i.e GO:0072766 centromere clustering at the mitotic interphase nuclear envelope where specific genes enable the clustering process.

Of course this might be different for higher eukaryotes.... so I would be happy with the above clarification...

[ValWood]

Sorry, this is a bit picky I'm a teeny bit precious about the fission yeast chromosome segregation and kinetochore annotation (accuracy and completion)

[hattrill]

Having a look at this https://academic.oup.com/nar/article/48/8/4161/5809156, looks like it may also be within an individual chromosome:"In Arabidopsis, similar to humans or mice, chromocenters are composed of the pericentric heterochromatin of individual chromosomes, which sometimes but not regularly associate".

Perhaps I have missed the kinetochore link. Can't see that there is one in the GO, but think that I am not understanding your point.

[ValWood]

I'm probably misunderstanding what it is (and obviusly most vertebrate centromeric regions are lots larger)

If a chromocentre can be a single chromosome (kinetochore plus pericentric region' how does that differ from "chromsome centromeric region (which includes the kinetochore and the pericentromeres).

I am more convinced that 'chromocentre' is not a fungal feature so I will request a taxon restriction for fungi.

[hattrill]

I haven't come across any examples in fungi, so I think that you'd be ok putting a taxon restriction there.

Think I'd have to have a look at the non-fly examples e.g. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4663263/ D.mel refer to bundles and the suggestion is that the composition differs from centromeric/pericentromeric regions. https://elifesciences.org/articles/34122

Where the line is drawn between higher-order domains/compartments/complexes/organisation is not easy to see here and most articles contain the word 'enigmatic'.

[hdrabkin]

So I'm good to Fix definition

A region in which centric, heterochromatic portions from more than one chromosomes form a compact structure.

Move to GO:0043232 | intracellular non-membrane-bounded organelle

?

[ValWood]

I think that would be really useful if everyone agrees.

[hattrill]

Agreed! Thanks @hdrabkin

[hdrabkin]

done