transcription factor activity

gocentral commented 21 years ago

The node 'transcription regulator ; GO:0030528' needs sorting out - importantly, we need to decide what to do about term 'transcription factor activity ; GO:0003700' - make obsolete? See email archive for more info.

Reported by: jl242

Original Ticket: "geneontology/ontology-requests/705":https://sourceforge.net/p/geneontology/ontology-requests/705

gocentral commented 21 years ago

Logged In: YES user_id=451873

This item on hold until the next GO Consortium meeting (3-4th June).

Original comment by: jl242

gocentral commented 20 years ago

Logged In: YES user_id=516865

It should really should go completely. hmm that will mean me reannotating the 30 odd TFs I went through today....rats

Original comment by: ValWood

gocentral commented 20 years ago

Logged In: YES user_id=451873

Was SF item [ 832370 ] transcription questions from Val:

GO:0003702 : RNA polymerase II transcription factor activity (and its child terms) and GO:0003709 : RNA polymerase III transcription factor activity

are compound terms of component and function.

The more appropriate terms already exist GO:0003700 :transcription factor activity, althought the appropriate relationships don't appear to all be represented.

I'd suggest getting rid of the former, and moving any of the children which are still required (prerhaps the specific and non-specific terms?) to under GO:0003700

I think you can get rid of enhancer binding as doesn't this have to bind promoter or enhacer (at least the complex) the by definition.

Original comment by: jl242

gocentral commented 20 years ago

Logged In: YES user_id=451873

Was SF item [ 831214 ] transcriptional repressor activity (def) from Val:

GO:0016564 transcriptional repressor activity no def

is a child of GO:0030528 transcription regulator activity def Plays a role in regulating transcription; may bind a promoter or enhancer DNA sequence or interact with a DNA-binding transcription factor.

however,

GO:0003714 transcription co-repressor activity The function of a transcription cofactor that represses transcription from a RNA polymerase II promoter; does not bind DNA itself.

I'm assuming that transcriptional repressor activity needs to physically bind DNA, can a def be added to say this

propose Plays a role in negatively regulating transcription by binding a promoter or enhancer DNA sequence.

Original comment by: jl242

gocentral commented 20 years ago

Logged In: YES user_id=516865

Hi Jane,

Re: Curator requests item #831214, so, if GO:0016564 doesn't need to bind DNA but can bind other TF's too....then

GO:0003714 transcription co-repressor activity should have GO:0016564 as a parent (or some ancestor) in the revamp (it doesn't at the moment), which is what prompted the original Q.

I also wasn't sure whether it is true that a transcription co-repressor ONLY repressed trranscription from RNA polII as stated in hte definition?

cheers

Val

Original comment by: ValWood

gocentral commented 20 years ago

Logged In: YES user_id=436423

We have (with defs):

-%transcription regulator activity ; GO:0030528 (superset - covers both DNA binding and non- DNA binding) --%transcription factor activity ; GO:0003700 (binds DNA) --%transcription cofactor activity ; GO:0003712 (doesn't bind DNA)

So the activators, repressors, coactivators and corepressors should be done the same way (term names abbreviated to save some typing):

-regulator GO:0030528 --factor GO:0003700 % DNA binding ---DNA binding coactivator (GO:new) % DNA binding ---DNA binding corepressor (GO:new) % DNA binding --cofactor GO:0003712 ---coactivator GO:0003713 --activator GO:0016563 ---coactivator GO:0003713 ---DNA binding coactivator (GO:new) % DNA binding --repressor GO:0016564 ---corepressor GO:0003714 ---DNA binding corepressor (GO:new) % DNA binding

(or with more new IDs if needed for other aspects of this item)

... and definitions added accordingly.

I don't think any of the above should be specific to any one polymerase, so actually probably ought to change the wording and placement of the existing coactivator and corepressor terms, and add new terms/ids for the generic ones. arg. m

Original comment by: mah11

gocentral commented 19 years ago

Logged In: YES user_id=546388

Proposed changes to transcription factor activity branch of FUNCTION

Karen Christie and I are concerned that the transcription factor activity term may be obsoleted, mostly because it appears to be ill-defined and/or its children misplaced, and/or people appear to be mis-annotating. There are currently over 5100 non-IEA annotations to this term or its children; obsoleting would involve a massive clean-up.

These changes are based on my discussions with Karen Sprague (U. Oregon), and Steve Buratowski's web page (http://tfiib.med.harvard.edu/ transcription/basaltx.html). Dr. Sprague voiced a concern that certainly people in the field think of something having transcription factor activity. Buratowski makes a really clear distinction between 'basal' and 'regulatory' transcription factors. Basically, the phrase 'transcription factor' cannot be used exclusively for the things that bind DNA and then regulate the transcription activity of a gene or group of genes. The transcription field also includes the 'basal transcription factors', not all of which bind DNA, as transcription factors. Buratowski uses the phrase 'regulatory transcription factor' to distinguish the ones that bind DNA and modulate the level of transcription, presumably by interacting with the 'basal transcription factors', though it's not clear at this point exactly how most regulatory txn factors interact with the basal machinery to affect txn levels.

The current GO tree (http://www.godatabase.org/cgi-bin/amigo/go.cgi? view=details&depth=1&query=3700)

has a TPV for the term 'transcription factor activity' to be a child of 'DNA binding'. For that matter, as the term 'transcription regulator activity' is currently defined, maybe it should be a child of 'transcription factor activity', which seems to have a broader definition.

Karen thought that the current higher level term 'transcription regulator activity' may really be a process term. 'regulatory transcription factor activity', as per Buratowski's site, is a reasonably well defined function. Was 'transcription regulator activity' is meant to be a synonym of 'transcription factor activity'?

Karen suggests that the best option, would be to move 'transcription factor activity; 3700' directly under 'transcription regulator activity ; 30528' and make the 3 terms 'RNA polymerase I (or II or III) transcription factor activity' into is_a children of 3700. It makes no sense at all for 3700 to be a sibling of 3 terms that represent subsets of 'transcription factor activity'. Keeping the term independent would provide a term for prokaryotic transcription factors. The one term that is currently a child of 3700 already has appropriate specific parentage, so it should just lose it's specific parentage directly under 3700 (if 3700 is moved). Again, 3700 cannot be a child of 'DNA binding', so that will need to be changed. It would be a vast improvement to move the 'transcription factor activity; 3700' term to be a child of the 'transcription regulator activity ; 30528' term and to remove the 'DNA binding' parentage from the 3700 term.

This should then remove any confusion in whether something binds the DNA directly or not as being a factor vs co-factor, etc.

Original comment by: hdrabkin

gocentral commented 17 years ago

Logged In: YES user_id=436423 Originator: NO

Hi,

This item has been open for a long time. Please comment to let us know whether you would like it to remain assigned to you, or would prefer it to be reassigned. (You don't necessarily have to work on it immediately if you keep it; we just need to know whether it's still on your list.)

Thanks, Midori & David Ontology development group managers

Original comment by: mah11

gocentral commented 17 years ago

Logged In: YES user_id=451873 Originator: YES

I'm going to unassign this one to myself, gladly ;)

Original comment by: jl242

gocentral commented 17 years ago

assigned_to: jl242 --> nobody

Original comment by: jl242

gocentral commented 17 years ago

assigned_to: nobody --> kchris

Original comment by: mah11

gocentral commented 16 years ago

Logged In: YES user_id=473890 Originator: NO

comments transferred from [ 1422502 ] Transcription factor related https://sourceforge.net/tracker/index.php?func=detail&aid=1422502&group\_id=36855&atid=440764

Submitted By: Valerie Wood - val_wood Date Submitted: 2006-02-02 06:01

This is related to the 'TF related' 14573 on the annotation tracker.

Basically I am trying to get all of my general and seqeunce specific TFs separated by function term, (ignoring for the moment whether these are bona fida functions...I currently have one foot in both camps!)

Anyway I think now I maybe should have been using

specific RNA polymerase II transcription factor activity and general ....

So probably the new terms M. suggested in 14573 already exist

Should 'specific RNA polymerase II transcription factor activity' and 'general...' have the parent transcription factor activity? currently they don't which is why it took me a while to locate them...

Followups:

Comments Date: 2006-11-10 13:38 Sender: kchris Logged In: YES user_id=473890

I think calling this term 'regulatory transcription factor activity' would be a great idea. Shortly after the confrontational 2003 GO meeting at Stanford, I went surfing around some web sites put up by transcription people. There was one by Steve Buratowski that used exactly that phrase, though I should point out that I can't find that site right now...

Anyway, I think this name makes it clearer what we are binning into this group.

-Karen

Date: 2006-07-03 04:40 Sender: val_wood Logged In: YES user_id=516865

As a follow up to this....

As this term excludes general transcription factors, perhaps we could rename "transcription factor activity" to "regulatory transcription factor" syn rTF and mak transcription factor activity a synonym.

I think this would be more in keeping with the current def of

Date: 2006-02-17 05:46 Sender: val_wood Logged In: YES user_id=516865

Hi Karen, you are probably correct that

"general RNA polymerase II transcription factor activity" should not have the parent "transcription factor activity" which is good wbecasue it means that I didn't make Sandra remove the Interpro mappings unnecisarily.

This would work for me as all of my 'specific sequence binding transcription factors' would be annotated to this term, and my general transcription factors wouldn't be. I am trying to make it possible to retreive these two sets seperately based on their function annotation.

Date: 2006-02-16 17:18 Sender: eurie Logged In: YES user_id=554670

Huh... We did have a big discussion about this definition. I posted the concensus definition a while back and it looks like it has disappeared?

Here's the URL for the previous item

http://sourceforge.net/tracker/index.php?func=detail&aid=1114890&group\_id=36855&atid=440764

Date: 2006-02-16 16:26 Sender: kchris Logged In: YES user_id=473890

Hi,

I'm not sure that the "general RNA polymerase II transcription factor activity" term should have parent under the term "transcription factor activity" because the current term has parentage under "DNA binding" and I'm not sure that all of the general transcription factors have DNA binding activity. It's possible that some may bind to another factor or the polymerase, but not to DNA directly. I'll have to check, or maybe I could just send an email to my PhD advisor. Unfortunately, the def doesn't really seem to clarify what the term is intended for. The current def of:

transcription factor activity:

Any activity required to initiate or regulate transcription; includes the actions of both gene regulatory proteins as well as the general transcription factors.

seems about as broad as the def for one of its parent terms:

transcription regulator activity:

Plays a role in regulating transcription; may bind a promoter or enhancer DNA sequence or interact with a DNA-binding transcription factor.

Did this definition ever get updated after the January 2004 GO meeting in which transcription factor was such a contentious issue?

-Karen

Date: 2006-02-02 06:03 Sender: val_wood Logged In: YES user_id=516865

and should these also have DNA binding as a parent.

THese are the gene products i am trying to group. Sequence specific DNA binding transcription factors.....

Original comment by: krchristie

gocentral commented 16 years ago

Logged In: YES user_id=473890 Originator: NO

comments transferred from: [ 1953394 ] mp:specific RNA polymerase II transcription fact https://sourceforge.net/tracker/index.php?func=detail&aid=1953394&group\_id=36855&atid=440764

Submitted By: Valerie Wood - val_wood Date Submitted: 2008-04-28 06:09

GO:0003704 : specific RNA polymerase II transcription factor activity

does not have paretn transcription factor activity

this might already be submitted in one of the trasncription entries but I can't see it....

Followups:

Comments Date: 2008-04-30 10:01 Sender: kchris Logged In: YES user_id=473890 Originator: NO

OK, I will include this in my transcription reorg. I've assigned the item to myself.

-Karen

Date: 2008-04-29 08:01 Sender: val_wood Logged In: YES user_id=516865 Originator: YES

If thats the case I have used it totally wrongly! I have used it for sequnece specific DNA binsing transcription factors which control the transcription of subsets of polII transcribed genes (like forkheads, leucine zippers etc)

for example, ace2, atf21....

just checked and this also seems to be the case for SGD, but i can't say whther tohis is the cae for other organsims

Date: 2008-04-29 08:00 Sender: val_wood Logged In: YES user_id=516865 Originator: YES

If thats the case I have used it totally wrongly! I have used it for sequnece specific DNA binsing transcription factors which control the transcription of subsets of polII transcribed genes (like forkheads, leucine zippers etc)

for example, ace2, atf21....

just checked and this also seems to be the case for SGD, but i can't say whther tohis is the cae for other organsims

Date: 2008-04-29 07:23 Sender: gomidori Logged In: YES user_id=436423 Originator: NO

It may be an oversight, or it may be because of that whole DNA-binding business (the GO 'transcription factor' term is defined to mean only things that bind DNA directly). I'll ask Karen if she can look at this along with the other transcription stuff.

m

Original comment by: krchristie

gocentral commented 16 years ago

Logged In: YES user_id=473890 Originator: NO

comments transferred from: [ 1999361 ] transcription factor activity, child problem

Submitted By: Valerie Wood - val_wood Date Submitted: 2008-06-21 02:51

because specific RNA polymerase II transcription factor activity is not a child of transcription factor activity

I have an odd situation that I have more genes annotated to specific RNA polymerase II transcription factor activity (90) than I do to transcription factor activity (87)

can this parentage be added in idvance of the big transcription overhaul?

Thanks

Val

Followups:

Comments Date: 2008-06-25 02:43 Sender: val_wood Logged In: YES user_id=516865 Originator: YES

Hi Karen,

The query is easy for anyone who can do SQL queries. I can do it for pombe using our query interface. I'll see if somebody can provide the syntax to do the query generally.

I guess what we need is

for each organism

i) the number of genes annotated to both specific RNA polympolymerase II transcription factor activity and transcription factor activity

ii) the LIST of genes annotated to specific RNA polympolymerase II transcription factor activity and NOT to transcription factor activity and vice versa.

iii) the number of genes annotated to specific RNA polympolymerase II transcription factor activity and to DNA binding

iv) the LIST of genes annotated to specific RNA polympolymerase II transcription factor activity and NOT to DNA binding

This would provide a starting point to assess how these have been used....

Val

Date: 2008-06-24 11:10 Sender: kchris Logged In: YES user_id=473890 Originator: NO

Val said: "It should be fairly easy to do queries to see if people have done this generally. This would suggest a merge of these 2 terms."

So Val, is this you volunteering to do some analysis like this to help me figure out the appropriate fate for the "transcription factor activity" term? ;)

-Karen

Date: 2008-06-24 02:47 Sender: val_wood Logged In: YES user_id=516865 Originator: YES

I just did another check and all of my gene products annotated to specific RNA polymerase II transcription factor activity are also annotated to DNA binding (although I suspect they should all be annotated to sequence-specific DNA binding eventually)

So, I don't have an example of a specific RNA polymerase II transcription factor activity which does not bind to DNA....and I can't think or a gene product which would be described as this but does not bind to DNA .....

Val

Date: 2008-06-24 02:40 Sender: val_wood Logged In: YES user_id=516865 Originator: YES

Hi karen,

I realise this is a daunting task!

In case it helps, I have used these 2 terms completeley analogously. so that every gene I have annotated to

transcription factor activity is annotated to specific RNA polympolymerase II transcription factor activity

and every gane annotated to specific RNA polympolymerase II transcription factor activity is annotated to transcription factor activity (except the 3 in the difference above, but i can make an anntoation to "transcription factor activity" for these.

So i have used these 2 to have exactly the same meaning (DNA binding sequence specific transcription factors).

It should be fairly easy to do queries to see if people have done this generally. This would suggest a merge of these 2 terms.

I suspect though, because of the different uses of 'transcription factor' in the community (i.e for these sequence specific transcription factors, and also for general transcription factors), that it might be better if the primary term 'transcription factor activity' did not exist, but that all the terms used to represent 'transcription factor' with the various meaings were very explicit and 'transcription factor' was a synonym, of each of these terms....

Val

Date: 2008-06-23 16:34 Sender: kchris Logged In: YES user_id=473890 Originator: NO

Midori is correct about the reason this parentage is not already there. Because adding this parentage would give the "DNA binding" ancestry to "specific RNA polymerase II transcription factor activity", I think we should not do this until we are sure that the term "specific RNA polymerase II transcription factor activity" SHOULD be a child of "DNA binding". Currently I'm not, and I really won't be until I get farther into this.

I'm sorry I'm so slow Val, but I've been a little intimidated by the magnitude of the problems in this area and have been having a hard time wrapping my head around the issue. I've gotten a wiki page started (in the current ontology projects section) and am starting by going through the SF items to figure out all the issues. The problem you mention here, i.e. the def and placement of "transcription factor activity" is probably the crux of the entire issue since it's def and placement allowed it to be used inconsistently.

-Karen

Date: 2008-06-23 08:52 Sender: val_wood Logged In: YES user_id=516865 Originator: YES

You might be right.

The way I have used both of these DNA binding is a requirement but

specific RNA polymerase II transcription factor activity

does not specificy this.

However, if it doesn't bind DNA it wouldn't be a transcription factor it would be a regulator....

so, not sure.

Date: 2008-06-23 08:29 Sender: gomidori Logged In: YES user_id=436423 Originator: NO

I wonder if the reason it's not already there is the DNA binding thing that featured in so many GO_meeting discussions of transcription factor terms ...

anyway, I'll wait until Karen has a chance to comment, but I have no objection ...

Original comment by: krchristie

gocentral commented 14 years ago

comment transferred from:

ID: 2925558 - specific RNA polymerase II transcription factor activity

Is it necessary for genes annotated to this term to bind DNA? (all of my annotation to this term do bind DNA because of the useof the term "transcription factor" in the term name.

Howver, this is not currently specificed in the definiiton.

If it is not necessary to bind DNA then I wondered why it is called a transcription factor rather than a transcription regulator?

Thanks

Val

Original comment by: krchristie

gocentral commented 14 years ago

comment transferred from:

ID: 2993145 - "RNA polymerase II transcription factor activity" parentage (by Mike Livstone)

Shouldn't GO:0003702 "RNA polymerase II transcription factor activity" be a child of GO:0003700 "transcription factor activity" and not a direct child of GO:0030528 "transcription regulator activity"?

(Submitted while curating PTHR11829 using PAINT)

Original comment by: krchristie