Relationship between "nodulation" and related terms

[gocentral]

"GO:0051819 induction of tumor, nodule, or growth in other organism involved in symbiotic interaction" and "GO:0009877 nodulation" are not related close to one another in the BP ontology, which might be correct, but I believe they require clarification in the form of usage comments.

The nodulation term (GO:0009877) is a child of "GO:0044403 symbiosis, encompassing mutualism through parasitism," whereas GO:0051819 is a child of "GO:0051817 modification of morphology or physiology of other organism involved in symbiotic interaction," which has a part_of relationship to GO:0044403. GO:0009877 is defined as "the formation of nitrogen-fixing root nodules on plant roots" and GO:0051819 is defined as "The process by which an organism causes the formation of an abnormal mass of cells in a second organism, where the two organisms are in a symbiotic interaction." I assume that the two terms are intended to mean different things, but I think that since they are superficially quite similar, comments on how to use them would be very helpful to users.

Would GO:0009877 be used to annotate both plant and symbiont gene products? I assume this is the case, since it mentions "nitrogen-fixing," which is performed by the microbial symbiont, as well as "root nodules on plant roots," which clearly indicates the plant. On the other hand, should "GO:0051819 induction of tumor, nodule, or growth in other organism involved in symbiotic interaction" be used to annotate a gene product from a symbiont only? If this is so, then can we update the comments of these two terms to reflect this?

I also wonder about the definition of "GO:0051819 induction of tumor, nodule, or growth in other organism involved in symbiotic interaction," in particular the phrase "abnormal mass of cells." (I am aware that issues concerning the use of "abnormal" are "normal" are not new to the GO.) I think that nodule formation, like many other structures, is quite a "normal" part of many symbioses, including from the perspective of the plant. Perhaps we could simply cut the word "abnormal" from the definition without needing to take more drastic measures.

Finally, I think adding the synonyms "nodule formation" and "nodule development" (as previously suggested by Jane) might facilitate searching for the proper terms to use. Currently a query with the term "nodule" does not result in the term "nodulation"! Adding some appropriate synonyms that contain the word "nodule" might help guide users to "nodulation" and prevent inappropriate usage of "induction of tumor, nodule, or growth in other organism involved in symbiotic interaction" when this is not the intended term.

Reported by: mchibuco

Original Ticket: geneontology/ontology-requests/7987

[gocentral]

Hi Marcus,

I suspect that the underlying problem is that at the time that 'nodulation' was added -- years before PAMGO started up, iirc -- none of us were really thinking about annotating plant gene products versus annotating symbiont gene products. I could perhaps go further and speculate that people were thinking more consciously of plant than symbiont gene products, but that's even shakier, a conjecture based solely on the fact that GO:0009877 was added by TAIR curators.

I'll ask some of the TAIR curators to look at this item and comment; if a consenus emerges I'm quite happy to do the editing to implement it, but that's about all I can offer, since I don't usually have much to do with any of the interactions-between-organisms terms.

best, Midori

Original comment by: mah11

[gocentral]

I've added the synonyms.

The nodulation term IS intended to be used for both parasite and plant, in fact it's the poster-term for the documentation!

http://www.geneontology.org/GO.annotation.conventions.shtml\#newTerms

It should have two children:

induction of nodule morphogenesis in host (for annotating symbiont genes) induction of nodule morphogenesis by symbiont (for annotating host genes)

the former would get the parentage "GO:0051817 modification of morphology or physiology of other organism involved in symbiotic interaction"

I agree about the use of the word 'abnormal' in the definition of GO:0051817 - I reckon we just take it out.

Original comment by: jl242

[gocentral]

• assigned_to: nobody --> jl242

Original comment by: jl242

[gocentral]

Thank you Midori and Jane for your insights. Midori, your speculation on the origins of the terms sounds reasonable enough. Jane, I agree with you that there should be child terms of nodulation. I wonder if "induction of nodule morphogenesis" is too restrictive, however; I am getting hung up on the word "induction." Does this only cover the beginnings of the process, or would it address events occurring throughout the process such as the bacterial infection threads and such? I also wonder about phenomena such as down-regulation of nodule morphogenesis, and whether this would be covered by this term. It would seem like a "regulates" term might be in order. Perhaps I should wait until we hear back from TAIR.

Original comment by: mchibuco

[gocentral]

I too interpret that GO:0009877 can be used to annotate both plant and symbiont gene products while GO:0051819 appears to be more appropriate for annotating symbiont gene products only. I strongly support clarification in the comment field about term usage. This applies to a large number of interaction terms in addition to these two, one additional example is GO:0052033. I remembered that I had to ask the PAMGO people to clarify the usage of this term. The main reason is that these terms often have a very long name (generated in a certain format by the PAMGO group), some are hard to interpret. Clarification of term usage (host or symbiont) will be very useful to annotators.

Back to 0009877, I’m OK with the suggestion for two child terms specifying the contribution of host/symbiont gene products. I also understand Marcus’ concern about the word ‘induction’ implying the beginning of the nodule morphogenesis process. How about:

nodule morphogenesis regulated/modulated (other suggestion?) by host nodule morphogenesis regulated/modulated (other suggestion?) by symbiont

I favor keeping the current term ‘nodulation’ even we create these more specific terms.

I found the following suggestion a bit confusing (not the use of induction): I would interpret the ‘by symbiont’ term is to be used to annotate symbiont gene products. This highlights the need of usage clarification for these interaction terms.

[induction of nodule morphogenesis in host (for annotating symbiont genes)] [induction of nodule morphogenesis by symbiont (for annotating host genes)]

Donghui

Original comment by: donghui

[gocentral]

• status: open --> pending

Original comment by: jl242

[gocentral]

Hi Marcus - just revisiting this in a SF clear-up, sorry it's taken so long to resolve. I agree these terms are very confusing, and agree there should be some relationship between 'induction of tumor, nodule, or growth in other organism involved in symbiotic interaction' and 'nodulation'.

I'm wondering whether we might have to re-think the ontology structure for these terms...the original idea was for each symbiotic process to have a generic term which could be used to annotate both host and symbiont gps, with child terms reflecting the host and symbiont contribution. In practice, this has been quite difficult to orchestrate and has lead to some quite confusing term arrangements.

For the viral terms, we've ended up making terms that can be used to annotate either host or symbiont genes, and we'll capture whether it's a host or symbiont gene in the annotation itself. I wonder if that might not be a better way to handle all the terms in this node?

Then you could just have something like:

formation (development?) of symbiotic structure ---[i] nodulation ------[p] induction of nodule formation ------[p] nodule morphoenesis ------[p] entry of symbiont into developing nodule

and you wouldn't need to worry about which organism was allowed to use which term.

Anyhow, in the meantime I've removed the word 'abnormal' from those definitions.

Original comment by: jl242

[gocentral]

• assigned_to: Jane Lomax --> David Osumi-Sutherland

Original comment by: paolaroncaglia

[gocentral]

Hi, sorry this fell through the cracks due to a malfunctioning of the SF tracker. Assigning to David OS as he inherited symbiont-related work from Jane after she left. David, could you please see if the issues above are still current, and if anything needs done.

Thanks,

Paola

Original comment by: paolaroncaglia

[pgaudet]

Discussed with @thomaspd

Decisions:

1. Keep processes for symbiont genes and host genes distinct: nodulation = host genes induction by symbiont of nodulation, tumor, or growth in host = to annotate symbiont genes

2. Merge:
   'induction of tumor, nodule, or growth in other organism involved in symbiotic interaction' 'induction by symbiont in host of tumor, nodule, or growth' 'induction by symbiont in host of tumor, nodule, or growth containing transformed cells' as these terms all refer to very similar concepts (the definitions were essentially identical), and there is no reason to distinguish between these.
   Renamed "induction by symbiont of nodulation, tumor, or growth in host"

3. Change definition: "The process in which a symbiont causes the formation of a mass of cells in a host organism."

4. Added axiom: 'positively regulates' some nodulation

See also changes in #14807

Pascale