search

geneontology / go-ontology

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

placement of "sperm motility" #10905

Closed gocentral closed 8 years ago

gocentral commented 10 years ago

Hi,

I am looking at "sperm motility", which is an is_a child of "cell motility".

I think "sperm motility" should be moved down to be an is_a child of "ciliary cell motility" since the sperm flagellum is a specialized type of cilium. Then, the "regulation of sperm motility" terms should be be made types of the more general "regulation of ciliary cell motility" terms.

I also see that there is a term for "cilium movement involved in cell motility", but I don't see a parallel term for "sperm flagellum movement involved in sperm motility". It seems potentially useful given that there are a number of specialized proteins present in sperm flagella, but not in other cilia that contribute to the extreme needs of the sperm flagellum. Of course, it might be worth considering the relations to terms like "sperm capacitation" as well before making changes based on this last suggestion.

thanks,

-Karen

Reported by: krchristie

Original Ticket: geneontology/ontology-requests/10719

gocentral commented 10 years ago

Original comment by: jl242

gocentral commented 10 years ago

I added has_plasma_membrane_part some cilium to 'sperm cell'. Everything else should follow automaticallly once cl_imports.owl updates. https://code.google.com/p/cell-ontology/source/detail?r=346

Original comment by: cmungall

gocentral commented 10 years ago

grr, sourceforge lost my response

ok, cl_imports now has the link, but we need to fix the def of ciliary cell motility, apply:

https://www.ebi.ac.uk/panda/jira/browse/GO-232

Original comment by: cmungall

gocentral commented 10 years ago

Original comment by: paolaroncaglia

gocentral commented 10 years ago

(Jane, taking this one as it's related to cilia/flagella...)

Original comment by: paolaroncaglia

gocentral commented 10 years ago

Hi Chris,

Back to your comments: "I added has_plasma_membrane_part some cilium to 'sperm cell'. Everything else should follow automaticallly once cl_imports.owl updates." (dated March 18) and "ok, cl_imports now has the link, but we need to fix the def of ciliary cell motility" (dated March 20), on March 25th I edited 'ciliary cell motility' stemming from another ticket from Karen (https://sourceforge.net/p/geneontology/ontology-requests/10712/). The term is now 'cilium-dependent cell motility' (intersection of 'cilium movement' results_in 'cell motility'; but also is_a 'cilium or flagellum-dependent cell motility', has_part 'cilium movement'). I'm a bit confused now, unsure if the links added for SF 10712 are the most appropriate, and looking at https://www.ebi.ac.uk/panda/jira/browse/GO-232 doesn't help much. What is still missing to be able to infer the desired links? Also, when you say 'sperm cell', do you mean CL:0000019 'sperm'?

Thanks, Paola

Original comment by: paolaroncaglia

gocentral commented 10 years ago

Hi Karen,

FYI, this ticket generated a wider discussion. In short, we realized we need to review the whole node under 'cell motility'. This will take some time, but in the end all terms will be placed more appropriately, which was your request for 'sperm motility'. I hope you don't mind the delay; I'll label this ticket as a 'mini-project' as part of a strategy to allocate necessary time and resources to non-straightforward SF tickets. For reference, I'm copying below Chris' email response to my questions in the previous comment. We hope to discuss 'cell motility' in more detail during one of our up-coming editors calls.

Thanks, Paola

Hi Paola.

I think it's worth thinking about this one as a group.

How much time do you want to spend doing this right?

We could just assert the link and close the ticket

But I think it's worth looking at the mix of asserted children here and trying to approach this differently

is_a GO:0044763 ! single-organism cellular process is_a GO:0006928 ! cellular component movement is_a GO:0048870 ! cell motility is_a GO:0001539 ! cilium or flagellum-dependent cell motility is_a GO:0003411 ! cell motility involved in camera-type eye morphogenesis is_a GO:0016477 ! cell migration is_a GO:0021805 ! cell movement involved in somal translocation is_a GO:0021814 ! cell motility involved in cerebral cortex radial glia guided migration is_a GO:0030317 ! sperm motility is_a GO:0043107 ! type IV pilus-dependent motility is_a GO:0070358 ! actin polymerization-dependent cell motility is_a GO:0071975 ! cell swimming is_a GO:0071976 ! cell gliding is_a GO:0090247 ! cell motility involved in somitogenic axis elongation is_a GO:0097230 ! cell motility in response to potassium ion is_a GO:0097231 ! cell motility in response to calcium ion

What are the different mechanisms of motility? Are they disjoint? Can something be 'cell swimming' and cilium-directed and migration at the same time?

It looks like the differentia are a mix of mechanism (e.g. cilium, pilus), context (involved in, response to) or a descriptive measure of the motion or immediate environment (swimming, gliding). Or migration which seems insufficiently differentiated.

On closer examination of the children, the last category seems to disappear and that these are actually cryptic mechanism descriptions too: gliding is achieved through secretion of adhesion molecules and 'swimming' is either the cytoskeleton contracting the cell or GO:0071977 which seems insufficiently differentiated from its parent

I'd prefer to ground things more mechanistically or contextually.

Once this is done it gives us a better way to answer questions about where 'sperm motility' can go. Are there any other possible mechanisms other than cilium? If not, then a hidden GCI is justified. Does this hold for any cell with an X as part, where X is a possible mechanism of motility?

Fixing the ontology retrospectively takes time, but we should take the opportunity to fix things as we go along.

Original comment by: paolaroncaglia

gocentral commented 10 years ago

Original comment by: paolaroncaglia

gocentral commented 10 years ago

Back to the original request, we can't do this because C. elegans sperm are amoeboid.

Original comment by: ukemi

gocentral commented 10 years ago

Point taken for "sperm motility" generally.

Would it be worth having some child terms for "flagellar sperm motility" and "amoeboid sperm motility" (or whatever phrasing is best) to be able to connect the different types of sperm motility with their specific mechanisms?

From what I've been annotating in mice, it would be nice to have sperm motility connected to ciliary/flagellar motility, since it's based on the flagellum, but with lots of extra sperm specific proteins.

-Karen

Original comment by: krchristie

gocentral commented 10 years ago

Makes sense.

When doing this though we have to make sure everyone 'pushed down' their annotations. Perhaps by marking the grouper class as being 'do not annotate directly'? Otherwise we have everything at different levels, not good for reliable querying.

On 16 May 2014, at 14:07, Karen Christie wrote:

Point taken for "sperm motility" generally.

Would it be worth having some child terms for "flagellar sperm motility" and "amoeboid sperm motility" (or whatever phrasing is best) to be able to connect the different types of sperm motility with their specific mechanisms?

From what I've been annotating in mice, it would be nice to have sperm motility connected to ciliary/flagellar motility, since it's based on the flagellum, but with lots of extra sperm specific proteins.

-Karen

\ [ontology-requests:#10719] placement of "sperm motility"**

Status: open Group: MGI Labels: Cilia/flagella Other term-related request mini-project Created: Mon Mar 17, 2014 04:53 PM UTC by Karen Christie Last Updated: Thu May 15, 2014 05:47 PM UTC Owner: Paola Roncaglia

Hi,

I am looking at "sperm motility", which is an is_a child of "cell motility".

I think "sperm motility" should be moved down to be an is_a child of "ciliary cell motility" since the sperm flagellum is a specialized type of cilium. Then, the "regulation of sperm motility" terms should be be made types of the more general "regulation of ciliary cell motility" terms.

I also see that there is a term for "cilium movement involved in cell motility", but I don't see a parallel term for "sperm flagellum movement involved in sperm motility". It seems potentially useful given that there are a number of specialized proteins present in sperm flagella, but not in other cilia that contribute to the extreme needs of the sperm flagellum. Of course, it might be worth considering the relations to terms like "sperm capacitation" as well before making changes based on this last suggestion.

thanks,

-Karen

Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/geneontology/ontology-requests/10719/

To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/

Original comment by: cmungall

krchristie commented 8 years ago

In an email discussion, Paola has pointed out that the cross-product definition of “sperm motility” is to a CL term that is a SubClassOf “ciliated cell”. Thus the logical definition of the term “sperm motility” is inconsistent with its position in the ontology. http://bioportal.bioontology.org/ontologies/CL?p=classes&conceptid=CL%3A0000019

As David H said earlier, C. elegans has amoeboid sperm, so within actively annotated model organisms, we have two types of sperm motility:

The existing logical definition would be correct for "flagellated sperm motility" and such a term looks like it should be a child of this term: "cilium-dependent cell motility (GO:0060285)”.

In the recent email discussion involving John Van Dam, it was noted that the sperm of angiosperms are not flagellated. However, from Dresselhaus et al. 2016. Fertilization Mechanisms in Flowering Plants (PMID:26859271), it looks like the sperm of angiosperms are not considered to be motile cells, so possibly don't need a type of "sperm motility" term.

-Karen

@paolaroncaglia @JohnvanDam @tberardini

tberardini commented 8 years ago

Yep, no motile sperms in angiosperms.

However, gymnosperms have flagellated and motile sperm cells. http://www.ncbi.nlm.nih.gov/pubmed/16736257

krchristie commented 8 years ago

Thanks Tanya, I think that the flagellated motile sperm of gymnosperms would be covered by a general term for "flagellated sperm motility" though. I'm not aware of anything that would necessitate a special term here.

tberardini commented 8 years ago

Yes, I thought that a 'flagellated sperm motility' term would work for gymnosperms too. Cool biology, eh?

krchristie commented 8 years ago

Hi Paola,

While you're working on cilia GH tickets, would you see if you can take a look and possibly finish this one too. thanks,

-Karen

@paolaroncaglia

paolaroncaglia commented 8 years ago

Hi @krchristie ,

Yes, I'm hoping to work more on ciliary processes this week, and the motility branch is in my to-do list :-)

Cheers,

Paola.

krchristie commented 8 years ago

Thanks @paolaroncaglia !!

paolaroncaglia commented 8 years ago

https://github.com/geneontology/go-ontology/issues/10898

paolaroncaglia commented 8 years ago

I have reviewed this thread. Looking at things more closely, CL:0000019 ’sperm’ is a descendant of ‘animal cell’, see http://www.ontobee.org/ontology/CL?iri=http://purl.obolibrary.org/obo/CL_0000019. So, even though CL:0000019 ’sperm’ is not formally taxon-restricted in CL (…neither is ‘animal cell’!), I think it’s safe to say that CL:0000019 ’sperm’ refers strictly to animal sperm. Not to plant sperms, so that takes care of the whole angiosperms/gymnosperms stuff. :-) As for animals that have sperms that are not ciliated, such as C. elegans, that’s also taken care of, because CL:0000019 ’sperm’ is_a ‘ciliated cell’, which worm sperms clearly aren’t. (Note that ‘ciliated cell’ is formally linked to GO ‘cilium’.) Therefore, any GO term that cross-references CL:0000019 ’sperm’, including ‘sperm motility’, cannot be used for plants, nor for any non-plant species that doesn’t have ciliated sperms.

This said, the pending issues are:

ukemi commented 8 years ago

I'm not sure I am completely comfortable with this. I believe some crustaceans also have sperm without flagella. It seems to me the CL term is too restrictive.

paolaroncaglia commented 8 years ago

Hi @ukemi. Uhm, the CL term identifies a sperm cell that is found in animals AND has cilia. That doesn't mean that animals can't have non-ciliated sperms - it's just saying that the particular instance CL:0000019 ’sperm’ is a ciliated animal sperm cell. It just so happens that CL doesn't have an instance for animal sperm cells that don't have cilia. Does that make sense? :-)

paolaroncaglia commented 8 years ago

@ukemi We could certainly suggest to CL that they add the missing term, as long as we can provide them PMIDs or ISBNs to support it. I was just saying that the GO terms referencing CL 'sperm' actually mean 'ciliated animal sperm' blahblah.

krchristie commented 8 years ago

Hi Paola,

I agree with David that it might be good if we could use a CL term that is less restrictive to Metazoa, but not because of crustaceans that have sperm without flagella, but because of gymnosperms that do have flagellated sperm. If it is possible to use a more general term that would allow for any flagellated/ciliated cell, I think that would be better. Basically, I think that the CL term xref for "flagellated sperm motility" should be to a term with as broad a taxonomic range as possible since the presence of cilia/flagella seems to be very ancient. However, if the CL doesn't go to a level that makes that possible, I am OK with a less general term, though at some point in the future if GO expands its mandate for organismal coverage, we might need to use a more general term.

I am also concerned about having this term mean "flagellated sperm motility", but NOT changing its name to say that since there are other kinds of sperm motility that do not involve flagella. I think this will lead to annotators using this term to annotate other kinds of sperm motility. I think we need a general term for "sperm motility" and then under it terms for sperm motility that depend on different mechanisms, e.g. "flagellated sperm motility", "amoeboid sperm motility", perhaps others.

Basically, I think we need something like this:

- sperm motility
-- flagellated sperm motility
-- amoeboid sperm motility

It seems to me that the existing term for "sperm motility" refers to "flagellated sperm motility, based on its existing cross products. However, whether it is better for the existing term "sperm motility" to remain the top level sperm (to avoid making any current annotations incorrect) or to make it more specific based on its current cross-products is an open question that I don't feel strongly about.

-Karen

vanaukenk commented 8 years ago

@krchristie suggestion sounds reasonable to me. I agree that a more general 'sperm motility' parent term with child terms referring to the different mechanisms might be best. It also seems that there may be some work to do on the CL wrt the different types of sperm and getting the parentage correct?

ukemi commented 8 years ago

I like Karen's plan as well.

paolaroncaglia commented 8 years ago

Thanks all. Then this becomes a question for CL first of all... I'm not sure what their bandwidth is at the moment, and what timeline this would imply. @cmungall , any idea? I can certainly create a ticket on their tracker, unless one of you would like to give it a shot.

krchristie commented 8 years ago

Do proposed changes in GO really depend on CL? It seems that we could make the changes I'm proposing in GO without waiting for CL, using the existing CL reference for the xref of the "flagellated sperm motility" term for now and changing that later if/when CL provides a more general term.

paolaroncaglia commented 8 years ago

That could do too, yes. I'll come back to this hopefully tomorrow.

cmungall commented 8 years ago

It sounds like there should be a ticket with two items should be opened in the CL tracker:

paolaroncaglia commented 8 years ago

To start with, I opened a ticket in the CL tracker, feel free to subscribe: https://github.com/obophenotype/cell-ontology/issues/437

paolaroncaglia commented 8 years ago

Back to Karen’s comment

“I think we need a general term for "sperm motility" and then under it terms for sperm motility that depend on different mechanisms, e.g. "flagellated sperm motility", "amoeboid sperm motility", perhaps others. Basically, I think we need something like this:

Thinking about the best strategy here, I’m looking at the manual experimental annotations to ‘sperm motility’ and its descendants, including regulation terms. There are 139, to 9 species:

Mus musculus
Drosophila melanogaster
Homo sapiens
Caenorhabditis elegans
Rattus norvegicus
Danio rerio
Coturnix coturnix japonica
Sus scrofa
Lytechinus variegatus

All have flagellated sperms except Caenorhabditis, which accounts for 7 annotations (6 WormBase + 1 UniProt). Therefore, I will

paolaroncaglia commented 8 years ago

Caenorhabditis_annotations.xlsx

Hi @vanaukenk,

(Please refer to the checklist in my previous comment for a summary) There are currently 6 WormBase annotations to ‘sperm motility’ and regulation thereof. (Note the terms have just been renamed to ‘flagellated sperm motility’ etc. but this name change may not have percolated yet.) Could you please rehouse those annotations (listed in the attached file) to the new ‘amoeboid sperm motility’ terms below as appropriate?

GO:0097723 amoeboid sperm motility GO:1905416 regulation of amoeboid sperm motility GO:1905417 negative regulation of amoeboid sperm motility GO:1905418 positive regulation of amoeboid sperm motility

Thanks,

Paola

paolaroncaglia commented 8 years ago

Note for self: emailed GOA about the single UniProt annotation, here's my email text for reference:

"Could you please move the following annotation UniProtKB P34370 cil-1 GO:1902093 PMID:19781942 IMP P Inositol polyphosphate 5-phosphatase I5P2_CAEEL|cil-1|C50C3.7 protein taxon:6239 20130712 UniProt from the current term GO:1902093 'positive regulation of sperm motility' to the new term GO:1905418 'positive regulation of amoeboid sperm motility'?"

Alice S. picked it up, but Protein2GO is still unavailable due to the data centre post-shutdown issues.

paolaroncaglia commented 8 years ago

Hi @krchristie ,

I just added GO:0097724 ‘sperm flagellum movement’ (is_a cilium movement), and renamed GO:0030317 as ‘flagellated sperm motility’. If you wish, you may now request ‘sperm flagellum movement involved in flagellated sperm motility’ via TG template. Thanks.

paolaroncaglia commented 8 years ago

Hi @cmungall and/or @dosumis,

(You may ignore the rest of this ticket) Would it be appropriate to add the following links please:

cilium movement has_participant cilium sperm flagellum movement has_participant sperm flagellum cilium-dependent cell motility has_participant ciliated cell

Or would occurs_in or mediated_by be more correct? Thanks!

paolaroncaglia commented 8 years ago

Note for self: The action items left to do for this ticket (see unchecked boxes in my comment dated June 23rd 2016) are of minor priority, so I’ll set this ticket to pending.

krchristie commented 8 years ago

Thanks very much Paola for your work in resolving this issue. I've requested the ‘sperm flagellum movement involved in flagellated sperm motility’ term via TG template as you suggested.

thanks!!

-Karen

paolaroncaglia commented 8 years ago

Note for self: Alice emailed me this morning: "I tried to correct the annotation as you requested, but it looks like the changes have not be implemented in QuickGO and it does not recognize the new term. I'll try next week again."

paolaroncaglia commented 8 years ago

Hi @cmungall and @dosumis ,

May I please double-check something with you. For the cilia-GO project, I’m looking into creating links between cilium movement BP terms and ciliary CC terms/ciliated cell CL terms. In particular between

1) cilium movement and cilium 2) sperm flagellum movement and sperm flagellum

I’m leaning towards using has_agent CC, e.g. cilium movement has_agent cilium. occurs_in CC seems reductive and incomplete (e.g. cilium movement occurs_in cilium), but would propagate annotations. has_agent cilium/sperm flagellum AND occurs_in CL:0000064 ciliated cell/CL 0000019 sperm seems more complete, but CL is currently pointing to animal cells at the moment and not considering plants. So this option may create TPV currently and until CL expands to new taxa and retains the general classes (see https://github.com/obophenotype/cell-ontology/issues/437). Trying to create a link at a lower level between cilium-dependent cell motility and ciliated cell would also incur in the same problem.

Should I just add cilium movement/sperm flagellum movement has_agent cilium/sperm flagellum and stop there for now?

Thanks!

dosumis commented 8 years ago

has_agent* is not wrong, so I don’t see a problem with adding. We already have links in the other direction - via capable_of_part_of (this should probably be changed to capable_of).

I think we need to ditch the CL link for now as will remain too taxon restricted for the foreseeable future.

On Oct 6, 2016, at 8:43 AM, paolaroncaglia <notifications@github.com mailto:notifications@github.com> wrote:

Hi @cmungall https://github.com/cmungall and @dosumis https://github.com/dosumis ,

May I please double-check something with you. For the cilia-GO project, I’m looking into creating links between cilium movement BP terms and ciliary CC terms/ciliated cell CL terms. In particular between

1) cilium movement and cilium 2) sperm flagellum movement and sperm flagellum

I’m leaning towards using has_agent CC, e.g. cilium movement has_agent cilium. occurs_in CC seems reductive and incomplete (e.g. cilium movement occurs_in cilium), but would propagate annotations. has_agent cilium/sperm flagellum AND occurs_in CL:0000064 ciliated cell/CL 0000019 sperm seems more complete, but CL is currently pointing to animal cells at the moment and not considering plants. So this option may create TPV currently and until CL expands to new taxa and retains the general classes (see obophenotype/cell-ontology#437 https://github.com/obophenotype/cell-ontology/issues/437). Trying to create a link at a lower level between cilium-dependent cell motility and ciliated cell would also incur in the same problem.

Should I just add cilium movement/sperm flagellum movement has_agent cilium/sperm flagellum and stop there for now?

Thanks!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/geneontology/go-ontology/issues/10905#issuecomment-251890013, or mute the thread https://github.com/notifications/unsubscribe-auth/AAG4x0gaM-GtL8rPmwS1OUhAfloHyJwlks5qxKaSgaJpZM4IbpTZ.

cmungall commented 8 years ago

On 6 Oct 2016, at 1:37, David Osumi-Sutherland wrote:

  • Chris: we have some cleanup work to do here, merging with ‘mediated_by’. Official label in RO not has_agent - revisit?

+1

cmungall commented 8 years ago

Is has_agent correct in the general case here?

Maybe this doesn't work for cilia, but think of a person riding a bicycle. The agent in the bicycle-movement process is the person, not the bicycle or the mereological sum of person+bike.

In this case the bicycle undergoes movement, and the movement is undergone_by bicycle. We don't have such a relation in RO, but there are a number of cases where this would be useful.

paolaroncaglia commented 8 years ago

Hi @cmungall and @dosumis,

I agree with Chris - the real agent in the movement of the cilium is one of its parts and not the whole cilium (the molecular motor, which has ‘effectors’, again ciliary parts). So none of the available solutions look satisfactory at the moment, and it’s not a priority, so I’ll drop this.

David suggests changing motile cilium capable of part of cilium movement to motile cilium capable of cilium movement which sounds correct for this particular case, but I thought the use of capable of was reserved to MFs?

dosumis commented 8 years ago

but I thought the use of capable of was reserved to MFs?

Nope. If a structure is capable of carrying out a whole BP or MF then capable_of applies. If it only capable of carry out a part then capable_of_part_of applies. (The relations are not GO specific: In VFB we use capable_of an capable_of_part_of to record the functions of neurons.)

paolaroncaglia commented 8 years ago

Thanks @dosumis , I updated that link.

paolaroncaglia commented 8 years ago

Only action item left:

paolaroncaglia commented 8 years ago

Now moved to https://github.com/geneontology/go-ontology/issues/12722. So, closing this one.