NTR: dsDNA loop binding

[gocentral]

Interacting selectively and non-covalently with DNA to bring separated binding sites into juxtaposition with the intervening dsDNA forming a loop.

These two PMIDs may be relevant based on their abstracts (need to read them, there are many others) http://www.ncbi.nlm.nih.gov/pubmed/15950878 http://www.ncbi.nlm.nih.gov/pubmed/23514145

This NTR should be a parent of GO:0090202 ! gene looping

isa GO:0043566!_structure-specific_DNA_bindingtair

Reported by: jimhu

Original Ticket: geneontology/ontology-requests/10819

[gocentral]

Original comment by: tberardini

[gocentral]

• Description has changed:

Diff:

--- old
+++ new
@@ -6,4 +6,4 @@

 This NTR should be a parent of GO:0090202 ! gene looping

-is_a GO:0043566_!_structure-specific_DNA_binding
+is_a GO:0043566_!_structure-specific_DNA_bindingtair

• assigned_to: Tanya Berardini

Original comment by: tberardini

[gocentral]

Hi Jim,

Your definition sounds more like 'dsDNA loop formation', instead of binding an already existing dsDNA loop.

Also, how does this part of the SF issue relate to the rest?

"isa GO:0043566!_structure-specific_DNA_bindingtair"

copy/paste weirdness?

Original comment by: tberardini

[gocentral]

Yes, I agree on formation Also, yes on copy paste weirdness.

Original comment by: jimhu

[gocentral]

Repeating my last comment, this time with a question at the end:

Your definition sounds more like 'dsDNA loop formation', instead of binding an already existing dsDNA loop.

Would you like to request 'dsDNA loop formation' instead? Or did you really want the term 'dsDNA loop binding'? In this case, the definition should be something more like: 'Interacting selectively and non-covalently with DNA containing a double-stranded loop.'

Original comment by: tberardini

[gocentral]

Binding vs formation is a slightly tough call. I originally suggested binding because I think of this working by the protein trapping a loop formed by spontaneous 3-d diffusion that brings the two sites into juxtaposition, while formation might incorrectly suggest a more active process. But it's also not like the DNA forms the loop and then the protein binds a preformed loop in one step. Usually, I believe you have binding to one or both sites individually followed by the collision and binding. So that argues for formation.

Sorry to be so wishy-washy.

Original comment by: jimhu

[gocentral]

Won't 'gene looping' work? "The formation and maintenance of DNA loops that juxtapose sequentially separated regions of RNA polymerase II-transcribed genes." Is it too specific with the RNA pol II bit?

Original comment by: tberardini

[gocentral]

Yes, the RNAPII part is too specific. Lac repressor does looping. So does bacteriophage lambda repressor (which is what the CACAO student was reading about). I also think it is better to separate the activity from the role in regulation of gene expression. I'm not sure, but I think that this must also happen in some site-specific recombination systems.

Original comment by: jimhu

[gocentral]

How's this:

dsDNA loop formation, GO:new --[i] gene looping, GO:0090202

chromosome organization, GO:exists --[i] dsDNA loop formation, GO:new ----[i] gene looping, GO:0090202 --[i] chromatin organization, GO:exists ----[i] gene looping, GO:0090202

def: The formation and maintenance of DNA loops that juxtapose separated chromosome regions.

Original comment by: tberardini

[gocentral]

OK except for

• I'm not sure about "chromosome regions" insofar as people might think it didn't apply to plasmids or other extrachromosomal elements
• There should be something to capture that it happens in cis. This would not be for the synapsis in meiosis, recombination or transvection.

How about:

def: The formation and maintenance of DNA loops that juxtapose separated regions on the same dsDNA molecule.

Original comment by: jimhu

[gocentral]

I like it! Will add soon. Thanks, Jim.

Original comment by: tberardini

[gocentral]

• labels: --> New Term Request
• status: open --> closed-fixed

Original comment by: tberardini

[gocentral]

Added:

[Term] +id: GO:0090579 +name: dsDNA loop formation +namespace: biological_process +def: "The formation and maintenance of DNA loops that juxtapose separated regions on the same dsDNA molecule." [GOC:jh, PMID:15950878] +is_a: GO:0051276 ! chromosome organization

parentage of 'gene looping' changed as described in discussion above.

Original comment by: tberardini