tpv; Ty element transposition

[gocentral]

Not sure that this should have DNA recombination as a parent.

transposition includes other processes (i.e the movement of a member of the Ty1 family of retrotransposons from one location to another in the genome...includes nucleocytoplasmic transport of transcripts, transcription and translation of GAG, POL, env etc etc....

Reported by: ValWood

Original Ticket: "geneontology/ontology-requests/2820":https://sourceforge.net/p/geneontology/ontology-requests/2820

[gocentral]

Logged In: YES user_id=451873

Yep - I agree. Would you like us to create a 'DNA recombination during Ty element transposition' or suchlike term when we delete this relationship?

Original comment by: jl242

[gocentral]

Logged In: YES user_id=516865

I wonder if terms for each of the steps in transposition would be useful....however not sure that we need each 'family' of transposition if the steps are the same? (we have ty1 element transposition, Ty2 element transpostion etc.) Most of the steps will probably be common (even for viral integration). I doubt we need separate terms for each kind of transposon....

Original comment by: ValWood

[gocentral]

• milestone: --> TermGenie

Original comment by: jl242

[gocentral]

Logged In: YES user_id=451873

Yes, I see what you mean. i think it would make sense to have terms for the different mechanisms of transposition i.e. retrotransposon transposition (e.g. Ty elements), class II transposon transposition (e.g. P-elements) and class III, MITE transposition. We could them mere the existing terms into these broader terms.

What do you think?

Original comment by: jl242

[gocentral]

• assigned_to: nobody --> jl242

Original comment by: jl242

[gocentral]

Logged In: YES user_id=516865

I think so but probably should check with SGD as they are the only ones to use the child terms so far. Only one gene uses thes terms, DBR1 which is annotated to ty1 and ty3 transposition

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=516865

can we up the priority of this one.

I need the term "retrotransposon transposition"

(from my survey....

I find the graphic interface of amiGO to be very difficult to use. I really have never found it to be useful. It also has mistakes in process names. For example it refers to the transposons as Ty1 not the true name Tf1.

I need a general 'retrotransposon transposition' term to fix this. I'm not sure we need to have a term for each kind of transposon.

Val

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=516865

can we up the priority of this one.

I need the term "retrotransposon transposition"

(from my survey....

I find the graphic interface of amiGO to be very difficult to use. I really have never found it to be useful. It also has mistakes in process names. For example it refers to the transposons as Ty1 not the true name Tf1.

I need a general 'retrotransposon transposition' term to fix this. I'm not sure we need to have a term for each kind of transposon.

Val

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=473890

Hi,

I'll forward this to SGD to get someone to take a look at how this impacts us.

cheers,

-Karen

Original comment by: krchristie

[gocentral]

Logged In: YES user_id=1510062

Hi, here are my thoughts on this issue (Please bear with me as I have quite a few of them).

I do not think that this is a tvp. The definition of DNA recombination is "The processes by which a new genotype is formed by reassortment of genes resulting in gene combinations different from those that were present in the parents..." DNA recombination during transposition is not a particular step, but the larger goal of the mobile element. This would make a term "DNA recombination during Ty element transposition" equivalent to saying "DNA recombination during Ty DNA recombination". If you are thinking specifically of the combining of the mobile element DNA with the host DNA, that process is better described by the term "integration" as opposed to "recombination".

On to the next issue. It's no problem for the SGD gene(s) to be merged into "retrotransposon transposition". I think it would be the most logical if the current term "DNA transposition" under "DNA recombination" could be renamed to just "transposition". In the mobile element community, DNA transposition specifically refers to class II transposition. Under the "transposition" term can be the children "retrotransposon transposition" (exact synonym: Class I transposition, narrow synonyms: Ty element transposition, etc.), "DNA transposition" (es: Class II transposition, ns: P-element transposition, Tc1/mariner transposition, etc.), and "MITE transposition" (es: Class III transposition).

For each of these 3 classes, the basic mechanism of mobilization is similar and so you don't need to have additional terms for each specific type of mobile element. I also don't think it's necessary to have terms for the individual steps of transposition.

Regards, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=516865

Hi Julie,

First I have a question...

is transposition generally used to refer only to the i)integration step, or to the ii) entire transposon life cycle?

If it is i) then we need a broader term to cover the other steps of 'the retro element lifecycle'

At the moemtn I have made annotations where I have assumed i) and these would need to be moved to more appropriate terms.

If ii) is correct I still think that the recombination parent is a TVP....maybe it could just be removed and replaced by an integration child term.

Here is why.... I have genes which are involved in nucleocytoplasmic transport which are shown to be involved in 'retroelement transposition/retorelement life cycle' (i.e if you knock them out you interfere with the nuclear import step of the transpostion process)

This means if I annotate these genes to a 'retroelement transposition term' they are automatically annotated to DNA recombination, which is wrong.

I'm not an expert on this, but intuitively I wouldn't think of 'DNA recombination' as a 'goal of the mobile element' rather a step in , or outcome of the process.

I agree with the changes you suggest to the DNA transposition terms and children.

I still think it is necessary for terms for the individual steps (processes) involved in transposition/transposon life cycle, as I have genes which need these terms to be fully annotated.

What do you think?

Val

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1510062

Hi Val,

Transposition normally refers to ii) the entire transposon life cycle. Unless a mobile element has completed all the steps of the process, it is not considered to have undergone transposition.

The way people who study transposable elements think of this is that mobile elements have their own agenda, which pretty much is "I have to get from here to there". DNA recombination really is the overall process, and everything else is just to help it get there. Hence why I don't consider this is a tpv. If these gene products are involved in the life cycle of the retrotransposon, then they are involved in DNA recombination because they are a process that contributes to the reordering of the genome.

I think having part-of relationships for all of the steps of transposition is fine. I didn't realize that there was need.

So then that part of the ontology would look something like: DNA Recombination -(i)Transposition --(i)Retrotransposon Transposition ---(p)Transcription during retrotransposon transposition (and also place as is-a under "transcription") ----(i)Transcription during ret. tpn, RNA-dependent ---(p)Nucleocytoplasmic transport during ret. tpn. (and also under transport etc.) ---(p)Translation during ret. tpn. ---(p)Particle assembly during ret. tpn. ---(p)Integration during ret. tpn. --(i)DNA Transposition ---(p)mobile element binding during DNA tpn. ---(p)mobile element excision during DNA tpn. ---(p)Target site selection during DNA tpn. ---(p)Integration during DNA tpn. --(i)MITE Transposition

Your thoughts?

Regards, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=516865

Hi Julie,

In that case,I still think there is a problem here, for the reasons pointed out below. Could you show Eurie this thread, she will know what I mean. The specific example below is a protein involved in nucleocytoplasmic transport (nucleoporin nup124) , but NOT involved in recombination. Currently it is annotated to recombination because of this parentage.

The node needs refining, so that ONLY the steps of the 'transposition reaction' which are part of recombination are under this node.

Cheers

Val

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1510062

Hi Val,

I still stand by the opinion that all the steps of the transposition reaction are part of DNA recombination in this case, as the definition is written. I've forwarded on the thread to Eurie and asked her to comment.

regards, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=554670

I talked with Julie about this. What it boils down to is the absence of "univocity" when talking about recombination.

There is recombination as the transposon community views it, which is inclusive of the different aspects required to start and finish the tranposition event. And there is recombination as the DNA repair community views it, which involves the processes that affect the strands of DNA. It's this latter community who might find it odd if NUP142 showed up on the list. But the DNA repair community would want to see the integrases (MuA, MuB, etc.) on the list of genes that actually do the breaking/rejoining of the DNA itself. So we have to relate the two communities.

The current definition of "DNA recombination" is broad enough to cover both communities. But, we should be specific about the recombination that involves just the metabolism of the DNA.

So this is our proposal:

1. Change term name of GO:0006310 to "genetic recombination", which is already used in the definition

2. Add new term "DNA recombination" whose definition could be: Processes that involved breaking and rejoining DNA strands. [Or something totally better]

3. Move terms like "recombinational repair", "meiotic recombination", "meiotic gene conversion", etc. to be the child of "DNA recombination"

4. Make changes to tranposition branch as Julie recommends

The rough structure would be:

genetic recombination (GO:0006310) --(i) DNA recombination (GO:NEW) ----(i) meiotic recombination ----(i) recombinational repair ----(i) mitotic recombination ----(i) etc. with current terms --(i) tranposition (GO:0006313) ----(i) DNA tranposition

This would then allow any part_of terms of retrotranposition that are breaking/joining of DNA strands to have additional parentage. As well as terms like "transport" and "transcription".

does that work?

Original comment by: eurie

[gocentral]

Logged In: YES user_id=516865

Hi Eurie, Julie,

Sorry still confused here!

From the structure you suggest, my nup annotation will still have a parent genetic recombination, which would still be incorrect.

There seem to be incionsistencies in what is said below.

If "transposition" is "the entire transposon life cycle. Unless a mobile element has completed all the steps of the process, it is not considered to have undergone transposition"

If this is true recombination cannot be a parent, it is a part of the complete process which includes a number of different steps.

To me it seems like the arrangement should be:

recombination --recombination during transposition

transposition --transcription during retrotransposon transposition --transcription during ret. tpn, RNA-dependent --nucleocytoplasmic transport during ret. tpn.
--translation during ret. tpn. --recombination during transposition

With the alternative arrangement suggested below, all of these new transposition steps will have the recombination parent, which would compound the problem.

If this is really a case of the same word being used to describe two different processes then we need differnt terms to describe them.

Have you got a refernce which explicitly states that "all the steps of the transposition reaction are part of DNA recombination". Maybe we could use this as a starting point to make the new specific terms which are required. Is there another term perhaps which would be less confusing? I have looked around at transposon related sites using Google but I can't find anything which states or implies that 'transposition is part of recombination. If you can point me to something like this it might help me to understand a bit more fully.

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=516865

Also the same applies to

'viral life cycle'

We would not consider all of the viral life cycle as a type of 'recombination' although there is a recombination step in its integreation. This is analogous...

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1109706

Hi all,

I just had a conversation with Julie about this too. I think I see what's bugging me about the placement of DNA recombination. "DNA recombination" (and the broader "Genetic recombination" proposed with the same definition) as it is currently defined is too broad a term to exist as a child of "metabolism". As I read its current definition, "DNA (Genetic) recombination" includes chromosome assortment and non-integrating bacterial plasmid transformation, among other processes. Inclusion of these in the "recombination" definition creates a TPV, in my opinion, since no bonds between deoxyribonucleotides or base modifications have to be created or broken in either case, hence no DNA metabolism. In those two cases, the DNA molecules just get moved around, which qualifies as "transport", not "metabolism".

The term "viral genome transport in host cell" (GO:0046796) is an analogous term to what Val wants for transposons, I think, and that one is under the "transport" branch, not the "metabolism" branch. I think it would be a mistake for that term to have "metabolism" as a parent. It's a slippery slope from there to having most "[substrate] transport" terms as children of "[substrate] metabolism" terms, which I don't think is intended. To dig up a related example, "snoRNA localization" is a child of "snoRNA metabolism", but "snRNA transport" (GO:0006408) is not a child of "snRNA metabolism". So it seems there is some indecision about how transport and metabolism should be related.

I do think it would be useful to have a high level term that could unite the processes listed in the current "DNA recombination" term, like plasmid transformation, transposition, chromosome assortment, etc. How about something akin to "chromosome organization and biogenesis", over in the "cellular physiology" branch? Maybe "genome organization and biogenesis"?
This could leave a simple "DNA strand exchange"-type term under DNA metabolism, and allow the non-metabolic terms to be moved somewhere more appropriate.

Nick

Original comment by: nastover

[gocentral]

Logged In: YES user_id=516865

I was thinking about this more to.

In each case there is the broad process and the specific recombination/integration event.

1) The very broad 'transposition' (transposon life cycle)

and

2) the 'transposition reaction' (recombination event)

Which is essentially the same as Nick is saying. The placement of 1 under metabolism is wrong. (I should have read the definition of recombination before!)

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1510062

So, although Nick brought up a lot of other examples and issues it might be better to focus back on transposition itself.

The definition of DNA metabolism includes both "The chemical reactions and pathways involving DNA..." This definition encompasses both the broad and the specific. Transposition is a series of processes that does involve the breaking and formation of DNA bonds and thus does belong under DNA metabolism as defined broadly. If people want to separate out the individual reaction from the pathways portion, then I think that should be a separate discussion.

Val, what you proposed below is exactly what Eurie and I proposed with the addition of a parent term "genetic recombination" as a parent of both "(DNA) recombination" and "transposition(al recombination)". In this case, the parent term is the broad process and the specific event is captured with "DNA recombination".

I view the transport steps of retrotransposition analogous to "DNA recombinase assembly" being a descendent of "homologous recombination" and "DNA metabolism". Transport and recombinase assembly are both subprocesses of a larger process that results in the yet broader processes of DNA reordering and DNA bond breaking and making.

regards, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=1510062

Nick has suggested that the definition of "transposition" should be revised so I'd like to propose the following:

transposition: the processes involved in mediating the movement of discrete segments of DNA between nonhomolgous sites

for class I and II child nodes add: "via an RNA/DNA intermediate" for the class III child substitute "miniature inverted-repeat transposable elements" for "discrete segments of DNA"

Original comment by: juliep

[gocentral]

Logged In: YES user_id=1109706

I've mulled this over too long, and this branch might as well stay as it is. (though we may want to make some definition changes). Either under "DNA metabolism" or under something higher, I'm envisioning a bunch of child terms under "transposition" too, like transcription of retrotransposon, transport of transposon, targeting of transposon, excision/insertion of transposon, translation of transposase, etc. that should get dual parentage.
For that reason it makes more sense to me to stick the Rube Goldberg machine that is transposition (especially retrotransposition) off on its own, similar to "viral life cycle", and assign specifc "metabolism" and "transport" parentage to certain steps. But that's just the way I envision the ontology, and odd as it looks to me now (as it did to Val), Julie's convinced me the current version is internally consistent.

This item has brought up some other issues with the definitions that I don't know if we want to deal with now. Up to you if you're satisfied with the status quo, Val.

Nick

Original comment by: nastover

[gocentral]

Logged In: YES user_id=516865

Hmm, I'm still not convinced that there isn't something majorly wrong, but I don't know enough about the processes to pinpoint it but I'm sure its a terminology issue. It doesn't resolve my original probelm that I have nucler pore proteins annotated to DNA recombination and this can't be correct......Somehow the process for transposition seems to be broader than its general recombination parent.

Sorry to keep going on about this, can anybody else see what I mean?

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1510062

I do understand why intuitively it just seems weird to put transport terms under DNA metabolism.

However, I also think that the processes that contribute to the overall result of the transposon moving to a new locus are okay in that part of the tree. The basic process of transposition is moving a piece of DNA from locus A to locus B. This essentially is a form of DNA recombination. Different transposons have different paths of getting from A to B, and is where things become complicated. So now the question becomes where do we put the 'A to B path', under transposition or somewhere else in the ontology?

After discussing things more with Karen and Nick, where to put contributing-to-a-given-end-but-otherwise-unrelated processes seems to be a philosophical issue to post to the annotation list. Would it be alright to table transposition until the larger issue is resolved?

regards, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=516865

yes certainly! I'm still puzzling over it!

Val

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1510062

Hi Val,

I'm sorry that it's been a while--things have been rather crazy here as of late.

I'm still planning to place the issue of contributing processes before the annotation list but I've been thinking more about the transport part specifically. Would you mind sending the PMID of the paper you're planning to annotate from?

Thanks!

regards, Julie

Your thoughts?

regards, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=516865

Hi Julie,

Sorry I also meant to post an update but was busy. I was thinking about this some more, and i think this a fundamental issue in GO analogous to teh viral host/parent issue.

the problem is essentailly, that endogenous proteins which help the transposon to recombine (i.e. transposon life cycle), are not involved in recombiantion for the 'host'.

We need to avoid a situation where every 'host' gene which is somehow involved in transposition (i.e transcription, translation, transport etc) are not annotated to recombination (which is the current situation).

The paper which I annotated (quite a while back) was PMID: 10409764

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1510062

Thanks Val!

I absolutely agree that for viruses we want to keep the host/virus annotations separate.

However, I would argue that in the case of transposition, transposons are part of the 'host' organism and that by definition endogenous proteins involved in transposition are involved in recombination for that organism.

I do agree that transposition, and especially retrotransposition, are slightly strange beasts and it's not totally clear where to place it in the ontology but I would consider them separate from viruses even though mechanistically some are similar.

I'll go and solicit opinions from some experts in the transposon/retrovirus/DNA repair fields and see whether it would make more sense to them to categorize transposons as part of the host or their own entity.

take care, Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=1510062

Here's another proposal that hopefully will work for everyone.

What do we think of: Leaving DNA transposition where it is, but rename it "transposition, DNA-mediated" for class II transposition, and creating a new process term "transposition" under "cellular physiological process". This should enable the annotation of the transport terms without causing TPVs. The proposed structure and definitions for the new terms are below. The different classes of transposition will be grouped under here and given dual parentage where appropriate.

I am also leaning against adding any part_of process terms for the individual steps of transposition (with one exception). Because the nuclear pore proteins most likely have a broader substrate range than just the retrotransposon capsid, it would be more appripriate to give those gene products dual annotations, one under "DNA transport" and the other to "transposition, RNA-mediated". I'm making an exception for reverse transcription, where the enzyme is encoded by the transposon and specific for just the transposon.

cellular physiological process; GO:50875 .is_a transposition; GO:new ..is_a transposition, RNA-mediated; GO:new ...part_of transcription during RNA-mediated transposition, RNA-dependent; GO:new ..is_a transposition, DNA-mediated; GO:6313 (rename) ..is_a MITE transposition

DNA recombination; GO:6310 .is_a transposition, DNA-mediated; GO:6313

transcription, RNA-dependent; GO:6410 .is_a transcription during RNA-mediated transposition, RNA-dependent; GO:new

Definitions and synonyms transposition: the processes involved in mediating the movement of discrete segments of DNA between nonhomolgous sites

transposition, RNA-mediated: the processes involved in a type of transposition which occurs via an RNA intermediate exact synonyms: Class I transposition, retrotransposon transposition, retrotransposition narrow synonyms: Ty element transposition, Tf transposition

transposition, DNA-mediated: the processes involved in a type of recombination where discrete segments of DNA move between nonhomologous sites via a DNA intermediate exact synonyms: Class II transposition, cut-and-paste transposition narrow synonyms: P-element transposition, Tc1/mariner transposition, Tc3 transposition

MITE transposition: the processes involved in the transposition of miniature inverted-repeat transposable elements (MITEs) exact synonyms: Class III transposition

transcription during RNA-mediated transposition, RNA-dependent: the synthesis of DNA from an RNA transposon intermediate exact synonym: reverse transcription during retrotransposition

Your thoughts?

-Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=516865

That would appear to work! Does anybody see any problem with this?

I'm happy to make dual annotations for the nuclear pore proteins for the reasons you suggest.

Original comment by: ValWood

[gocentral]

Logged In: YES user_id=1109706

Looks good to me Julie. I like how the terms aren't very ambiguous the way you've written them.

Nick

Original comment by: nastover

[gocentral]

Logged In: YES user_id=1510062

Great! Thanks all for your input.

Sorry, I forgot to add the reference for the terms/definitions earlier. I'd like to cite "Mobile DNA II" ISBN: 1555812090 for all of them.

If it's alright I'd also like to slightly reword the definition for "transposition, DNA-mediated" to: the processes involved in a type of transpositional recombination which occurs via a DNA intermediate

Regards, -Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=436423

OK, I think I got everything. New terms are:

transposition GO:0032196 transposition, RNA-mediated GO:0032197 MITE transposition GO:0032198 transcription during RNA-mediated transposition GO:0032199

I changed 'the processes' to 'any process' in the defs, because lately there's been a feeling that using the plural is ontologically dodgy (tho we haven't systematically cleaned up any except the development term defs).

If it's ok, we can close this.

m

Original comment by: mah11

[gocentral]

• assigned_to: jl242 --> gomidori
• status: open --> open-fixed

Original comment by: mah11

[gocentral]

Logged In: YES user_id=1510062

Looks good to me. Thanks Midori! (and thanks for taking care of intron homing as om/u)

-Julie

Original comment by: juliep

[gocentral]

Logged In: YES user_id=436423

OK, I'll close it. Thanks!

Original comment by: mah11

[gocentral]

• labels: --> Other term-related request
• status: open-fixed --> closed-fixed

Original comment by: mah11