Rename symbiosis and merge some children

pgaudet commented 6 years ago

Hello, The following terms are not processes; they correspond to modes of interactions between organisms, but not to processes.

'symbiosis, encompassing mutualism through parasitism' -> 'symbiont process'
mutualism -> annotations to a single paper PMID:24463512 -> symbiont process benefiting host
Merge: commensalism (0 annotations) into 'symbiont process'
Merge: parasitism into 'symbiont process' 5 annotations:
Updated definitions to be consistent with the process. -- symbiont process new def: A process carried out by symbiont gene products that enables a symbiotic interaction with a host organism. The various forms of symbiosis include parasitism, in which the association is disadvantageous or destructive to one of the organisms; mutualism, in which the association is advantageous, or often necessary to one or both and not harmful to either; and commensalism, in which one member of the association benefits while the other is not affected. However, mutualism, parasitism, and commensalism are often not discrete categories of interactions and should rather be perceived as a continuum of interaction ranging from parasitism to mutualism. In fact, the direction of a symbiotic interaction can change during the lifetime of the symbionts due to developmental changes as well as changes in the biotic/abiotic environment in which the interaction occurs. Microscopic symbionts are often referred to as endosymbionts.

--symbiont process benefiting host: new def: A process carried out by symbiont gene products that enables a symbiotic interaction with a host organism, that is beneficial to the host organism.

Gene	AC	Gene product	Assigned by	PMID	Date	TaxID	ECO
ENO	UniProtKB:Q27727	Enolase	CAFA	PMID:21949403	20171024	NCBITaxon:5833	IDA
BN1205_061250	UniProtKB:B9PW35	Fructose-bisphosphate aldolase	CAFA	PMID:19380114	20170123	NCBITaxon:432359	IMP
groL	UniProtKB:O66198	60 kDa chaperonin	CAFA	PMID:11333970	20160315	NCBITaxon:1028307	IMP
PB000456.03.0	UniProtKB:Q4YQJ5	Enolase, putative	CAFA	PMID:21949403	20171024	NCBITaxon:5823	IDA
mTXNPx	UniProtKB:Q95U89	Peroxidoxin	CAFA	PMID:22046130	20171025	NCBITaxon:5671	IMP

pgaudet commented 6 years ago

@pmasson55 Can you look at this ? See if you like the new definitions.

Thanks, Pascale

ValWood commented 6 years ago

Hi @pgaudet I am working with PHI-BAse to formalise their phenotype ontology terms. We were defining phenotypes and based on GO so it would be good to be tagged in any major changes like this then we can work out how to handle.

Tagging @CuzickA

pgaudet commented 6 years ago

Great, will do

pgaudet commented 6 years ago

In fact it would be nice to set up a phone conference with them to see if they have any feedback on this area of GO, in particular how to represent interactions - given that we annotate gene products, ie the perspective is always from a single specie's point of view, some terms could be formulated better.

Pascale

ValWood commented 6 years ago

Yes, I think a call will be required as soon as the ontology is formalized using GO.

Note that the pathogen community are using the word "interactions" here to describe the phenotype of a pathogen interaction with a host (for example a change in pathogenicity/virulence/mutualism).

CuzickA commented 6 years ago

Thanks for tagging me in this ticket. I have passed this information onto Kim Hammond-Kosack and she wanted me to feed back her thoughts to you....

tberardini commented 6 years ago

How about contacting someone who was involved in the PAMGO project - Candace? Trudy? That group spent a lot of time developing the original set of terms, including agonizing over what to call what eventually ended up as 'symbiosis, encompassing mutualism through parasitism.'

pgaudet commented 6 years ago

Hi @tberardini Do you have their contact info ?

pgaudet commented 6 years ago

@CuzickA

GO represents the function of gene products. Do you annotate lipids and other molecules ? I don't think the ontology supports that- Comments @ukemi @thomaspd @cmungall ?

tberardini commented 6 years ago

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2654661/

Has authors:

Trudy Torto-Alalibo Candace Collmer Michelle Gwinn Giglio

From #12613: (Diane Inglis attempted to contact PAMGO people.)

Candace responded that she is retired and the email to Trudy failed. Candace suggested Brett Tyler be contacted and our PAMGO info updated. I just got an out of town until 8/26 autoresponse from Brett's email address. Brett.Tyler@oregonstate.edu.

Michelle is still at IGR: http://www.medschool.umaryland.edu/profiles/Giglio-Michelle/

addiehl commented 6 years ago

I'm pretty sure Jane Lomax acted as GO editor for PAMGO, and may have knowledge to share. I don't have her Github ID or contact information, but perhaps someone does.

CuzickA commented 6 years ago

Tagging @KEHammond

dianeoinglis commented 6 years ago

Hi. Martin Urban with Phi-base was the best contact when we modified and developed some 'mycelium'-,related terms. He deals with host-fungal pathogen interactions, mainly in plants. We cc'd Brett Tyler to keep him abreast of the changes and receive his feedback.

dianeoinglis commented 6 years ago

I feel that leaving out specific mention of the host-symbiont interaction-type in the term names would be an unfortunate loss of information. Would people consider the following?...'symbiont parasitic process' which could be defined as a gene product that is used by a symbiont to its benefit at the expense of the host. The other related terms could be 'symbiont mutualistic process' and 'symbiont commensalistic process.'

cmungall commented 4 years ago

not re-opening, but noting that these is a paper out with some comments on the actions that were performed in this ticket

https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-020-01273-0 by @hoganwr's group

Unfortunately, we discovered that the definition of symbiont process (GO:0044403) in GO is also inadequate for reasons similar to the ones described already. GO defines symbiont process as “A process carried out by symbiont gene products that enables a symbiotic interaction with a host organism.” In addition to being circular, we found that the use of ‘host organism’ implies a narrower definition of ‘symbiosis’ than what de Bary intended, as it excludes symbiotic interactions that do not involve a host. For example, lichens are a composite of algae or cyanobacteria plus fungi and the fungi are not hosts for the algae or cyanobacteria.

Moreover, we discovered that GO had originally assigned a different term and definition to symbiont process (GO:0044403), which we actually found to be more appropriate:

‘symbiosis, encompassing mutualism through parasitism’ = def. - An interaction between two organisms living together in more or less intimate association.

However, the current release of GO now contains the new, circular definition, as well as two new classes—interaction with host (GO:0051701) and interaction with symbiont (GO:0051702)—which are both currently defined exactly the same as “An interaction between two organisms living together in more or less intimate association” (i.e., the previous definition for symbiosis, encompassing mutualism through parasitism (GO:0044403))

For these reasons, our re-use of GO classes was limited to multi-organism process (GO:0051704) and interspecies interaction between organisms (GO:0044419). We logged issues on the GO tracker highlighting the problems with these definitions. The GO team subsequently changed the definition of symbiont process (GO:0044403) to: “A process carried out by symbiont gene products that enables the interaction between two organisms living together in more or less intimate association.” Now symbiosis is not the interaction among two or more organisms of different species, but merely cellular processes that “enable” the interaction. This definition remains inadequate to our purposes as it excludes the interaction itself. Discussions with GO are ongoing.

This is a good paper (though I think the authors were not aware of the work by @mgiglio99 and others in https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2654661/).

I think some of the concerns are valid and may have been brought up by others in this ticket (and are reflected in the open ticket: #15497)

My understanding of the intent of the changes undertaken in this ticket is that GO is ceding the general concept of symbiosis. We are not the ontology people should use for general ecological concepts (despite the fact "biological process" sounds very broad, it's about genetically programmed processes). This is implicit in our use of a label "symbiotic process" and the relegation of the string "symbiosis" to a non-exact synonym.

In future, we need to do more coordination here. Many people were relying on GO for the general concept of symbiosis, and its children (which were merged in - these should have been obsoleted not merged). There may be other ontologies like ecocore willing to take over the general concept (cc @pbuttigieg @diatomsRcool). I think our move here has confused people.

I think in general the he idea of GO focusing on evolved processes carried out by gene products that are part of a larger multi-organism-level is a good one, but maybe this could have been done differently.

Perhaps the existing ID should have been obsoleted, as it has changed so much
We should have had a consider tag to an ID in a different ontology that was willing to do the job GO was previously doing
I'm not sure that suffixing "process" on the end of "symbiotic" is such a good way of differentiating what we mean from the general ecological concept.
The insertion of "gene products" into the def seems a bit odd, is this carried out by gene products any more or less than anything else in GO? Is the intent to make this a cellular process?
And given existing problems (#15497) it's not clear that the GO concept now has any ecological meaning and really in fact just means any interaction between organisms (sensu inclusive of viruses) where they are not the same species

mgiglio99 commented 4 years ago

Hi @cmungall As I mentioned in https://github.com/geneontology/go-ontology/issues/15497, the changes initially made in this issue should never have been done. I don't mean to sound harsh and believe me I know there were (and are) a lot of issues with a lot of the PAMGO terms, but when I saw what had happened to "symbiosis..." I had a small panic attack. During the time when "symbiont process" existed, things in this branch were even more messed up. Changing "symbiont process" to "symbiotic process" was one of the first things that happened in the multiorganism group. And we've continued to try to sort things out in that context and I think it has improved a lot from what it was then. I guess these authors didn't realize that "symbiont process" has since turned back into what it was before, but now framed as a process with "symbiotic process". It's too bad they didn't see this since I think they would have been happy.

Chris, I agree with your bulleted points at the end of your above post. In fact one of our recent multiorganism discussions was whether to just get rid of symbiotic process altogether in favor of just "interspecies interaction between organisms", which would follow directly from your last point. I've struggled with this as the original symbiosis term was near and dear to my heart. But I guess we just need to figure out at what level we want GO to be capturing these kinds of relationships and then we can proceed from there.

mgiglio99 commented 4 years ago

Tagging @nsuvarnaiari and @dsiegele so they can be in this thread.

cmungall commented 4 years ago

Thanks for your comments @mgiglio99. I wasn't aware of the full history (https://www.ebi.ac.uk/QuickGO/term/GO:0044403 - click change log).

I guess these authors didn't realize that "symbiont process" has since turned back into what it was before, but now framed as a process with "symbiotic process". It's too bad they didn't see this since I think they would have been happy.

I see. The authors' primary criticism was that the definition was circular, but this ceased to be the case on 2019-03-16. The paper only just came out - it would be great if authors/reviewers would check these things! But to be fair ontologies should not have this degree of churn on definitions!

In fact one of our recent multiorganism discussions was whether to just get rid of symbiotic process altogether in favor of just "interspecies interaction between organisms", which would follow directly from your last point.

I think this would be a shame. I worry the term has been used to inconsistently that any original intended distinction may be lost, but now I see your comments on the other ticket (#15497) I think the inconsistency may have a single source (poor KW mapping).

I have further thoughts but perhaps best articulated on one of the multispecies calls

I re-opened this ticket but not sure if it helps as we now have the other ticket....

hoganwr commented 4 years ago

In Matt's defense, he worked extremely hard to keep up and use the up-to-the-minute definitions. So the intent and attempt was absolutely there. IDK how often Ontobee loads the latest GO, but maybe that was the issue. He can speak for himself, but if I had to guess, I'd say he used Ontobee.

So, we absolutely 100% tried our level best. Not sure where the breakdown happened.

Plus, we had submitted this to ICBO last year, then submitted to JBSM, then JBSM special issue bombed, and given JBSM long TATs, retracted it from there and submitted to BMC Med Inf & Dec Mak. In between revisions, there was lots of GO changes.

On Mon, Oct 19, 2020 at 3:47 PM Chris Mungall notifications@github.com wrote:

Thanks for your comments @mgiglio99 https://github.com/mgiglio99. I wasn't aware of the full history ( https://www.ebi.ac.uk/QuickGO/term/GO:0044403 - click change log).

I guess these authors didn't realize that "symbiont process" has since turned back into what it was before, but now framed as a process with "symbiotic process". It's too bad they didn't see this since I think they would have been happy.

I see. The authors' primary criticism was that the definition was circular, but this ceased to be the case on 2019-03-16. The paper only just came out - it would be great if authors/reviewers would check these things! But to be fair ontologies should not have this degree of churn on definitions!

In fact one of our recent multiorganism discussions was whether to just get rid of symbiotic process altogether in favor of just "interspecies interaction between organisms", which would follow directly from your last point.

I think this would be a shame. I worry the term has been used to inconsistently that any original intended distinction may be lost, but now I see your comments on the other ticket (#15497 https://github.com/geneontology/go-ontology/issues/15497) I think the inconsistency may have a single source (poor KW mapping).

I have further thoughts but perhaps best articulated on one of the multispecies calls

I re-opened this ticket but not sure if it helps as we now have the other ticket....

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/geneontology/go-ontology/issues/14807#issuecomment-712402526, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJR55WRAUH6U3LZT64BFVDSLSJVVANCNFSM4EJIPKHQ .

ValWood commented 4 years ago

so is this now fixed?

geneontology / go-ontology

Rename symbiosis and merge some children #14807