Re: [Obo-discuss] IDO in GO

[Jane L. <ja...@eb...>]

Here is the poll for our phone call:

http://doodle.com/2b5g3i5b9giw97sv#table

cheers,

Jane

On 08/11/2013 01:27, Lindsay Cowell wrote:
> Dear Jane,
>
> Thank you for your email.
>
> On Nov 7, 2013, at 6:00 AM, Jane Lomax wrote:
>
>> Hi Lindsay
>>
>> We are hoping to use some IDO classes in GO logical definitions and 
>> annotations. This will probably be just symbiont and host to begin 
>> with, with possibly symbiont role, host role and some of the 
>> subclasses of symbiont being used further down the line.
>
> That's great.  And please let us know if the definitions work for you 
> or if you suggest revisions.  We tried to be consistent with the GO 
> process terms, so hopefully we are in good shape, but let us know if 
> some adjustment is needed.
>
> There are a lot of host and host role terms floating around, so it 
> would be great to have a concerted effort to reconcile these.
>
> Some uses are indeed more general than the infectious disease domain, 
> but we didn't find an existing term that we thought was appropriate so 
> included one in IDO.
>
> Most of the versions I have seen focus on "host" as used in the 
> context of symbiosis.  We defined the term more broadly, both to 
> accommodate usage in our domain but also to avoid clashing with other 
> domains.  An example of an alternative use is host as someone who has 
> received a transplant or medical implant of some kind.  Thus, we 
> defined host very generally, too generally some would say, but we 
> could think of counter examples for all narrower definitions we came 
> up with.
>
>>
>> What we really need, however, is a class for an organism that is the 
>> 'recipient' of a venom or toxin, where that organism is not a host. 
>> So for example the recipient of a snake bite, or bacterial toxin.
>
> Interesting that you raise this issue, as when we were first 
> developing IDO, disease caused by bacterial or fungal toxins when 
> there is no infection (the toxin is consumed but the bacteria or fungi 
> are not, or the bacteria/fungi are consumed but are not alive, or are 
> consumed but are cleared from the digestive track without establishing 
> an infection), caused some difficulty for us.  In the first two cases, 
> the affected organism is not a host to the organisms that produced the 
> toxin (but may be a host to the toxin, see below).
>
> I would have to re-read some of our definitions to remind myself how 
> we handled this, but I think we may have decided to ignore these.  It 
> can be argued that these are not infectious diseases, but they are 
> relevant for us because the source of the toxin is in many cases an 
> infectious agent.
>
>>
>> In GO we currently use 'other organism' which is unsatisfactory 
>> because, first, it /really/ isn't a cell component and second, 
>> because it doesn't communicate the role the organism is playing.
>>
>> Do you think this is in scope for IDO? Not sure what you might call 
>> the class - bitten organism? poisoned organism?
>
> We defined host role and host in a general enough way that more 
> specific types could accommodate the cases you mention.  It is open 
> for discussion of course whether people agree with this approach, but 
> as things currently stand this could be done relatively easily.
>
> Scope is an interesting question.  Many of the terms you will need 
> probably fall outside the scope IDO, but I don't know of another 
> ontology that seems more appropriate.  We could certainly include in 
> IDO the upper level terms, since we will need them as parents to some 
> terms that are within our scope.  We could also create a small 
> ontology specifically for organism roles (or something) that works for 
> all of our use cases.
>
> Perhaps we should arrange a call to discuss the best approach.
>
> Thanks,
>
> Lindsay
>
>>
>> thanks,
>>
>> Jane
>>
>
> -------------------------------------------------------------------
> Lindsay G. Cowell, PhD
> Division of Biomedical Informatics
> Department of Clinical Sciences
> University of Texas Southwestern Medical Center at Dallas
> 5323 Harry Hines Blvd.
> Dallas, TX  75390-9066
>
> F4.212b, MC9066
>
> phone: 214-648-2289
> fax: 214-648-2064
> Lin...@ut... 
> <mailto:Lin...@ut...>
>
>
> ------------------------------------------------------------------------
>
> UT Southwestern Medical Center
> The future of medicine, today.