Device utilizes material, and device capable of/enables process relationships

[ddooley]

RO has a shortcut "utilizes" relation (http://purl.obolibrary.org/obo/ro.owl/RO_0017001) between a device and a material: "X utilizes Y means X and Y are material entities, and X is capable of some process P that has input Y. , e.g. "A diagnostic testing device utilizes a specimen." It could also on the face of it then utilize assay consumables. utilizes has property chain "(capable of o has input)".

@bpeters42 mentioned there is an issue with some problem with a deprecated term or expression being used in OBI to connect a device to a process directly, and so should comment here about further action needed in OBI on this.

@ddooley thought RO needed a shortcut for a relation between devices and processes, but it looks like "capable of" and "enables" will fit the bill!

[wdduncan]

The following property chain is defined for RO:utilizes:

'capable of' o 'has input'

RO:enables doesn't seem quite right b/c it is the machine that carries out the function, not the material that is being analyzed (i.e., the material that is the input to process).

[ddooley]

@wdduncan "enables" is not quite right for what situation? (My side note was that I was looking for [device] capable of some [process], so its good for my use case, e.g. kettle enables some 'water boiling').

[wdduncan]

@ddooley Sorry if I misunderstood your comment. Our use case is:
[device] capable of some [process] and [that process] has input [X].

So, to use your example (nice example by the way), it would be:
kettle capable of some (boiling process that has input some water)

Or, to put it more simply:
kettle utilizes some water

[ddooley]

Ok, got that use-case and it will be useful for our FoodOn modelling too.

[addiehl]

The English language definition of utilizes: "to make use of : turn to practical use or account." [Merriam-Webster]

When I think of the relationship of the kettle to water, I think the kettle has the function to boil water. I utilize the kettle to boil water.

The PCR machine has the function to amplify DNA. I utilize the PCR machine to amplify DNA.

So I disagree with using utilizes as a relation in the manner as suggested above, as it is contrary to the standard English usage of this verb.

[wdduncan]

@addiehl It is the "make use of" part of the definition that is of most import. It seems to me at least that there is a sense in which a kettle "makes use of" water to perform the boiling process.

We would be happy to consider other labels for the relation. You can see some of the discussion we had about which label to use here.

[bpeters42]

I thought the discussion @wdduncan pointed to is very helpful. I think a very simple label like 'utilizes' is going to be confused, as we clearly have very different things in mind, and when I look for sentences using 'utilizes', all kinds of things come up. Chris Mungall proposed 'capable-of-has-input'. While that may be ugly, I think we will need labels like this that are longer to make distinctions clearer.

In OBI we had 'utilizes device' as the relationship between a planned process and a device. that is already a bit better than 'utilizes' by itself'. It could even be 'planned process utilizes device'.

For the relation between a device and its input, it could be 'device utilizes input'.

So:

'hammer' 'device utilizes input' 'nail'
'hammering' 'planned process utilizes device' 'hammer'

Essentially giving strong hints on domains and ranges as part of the relation label.

• Bjoern

[wdduncan]

Thanks for you input @bpeters42!
I'm not opposed to changing the label (that is easy to do). But, I am opposed to changing it to capable-of-has-input (or something similar). I think will make the term too difficult for non-ontologists.

I understand your point about 'utilizes' being very general. We can be make it little more refined with labels like:

• device utilizes
• device utilizes material
• device uses (material)
• device processes (material)

I would prefer not to use a label like device utilizes input b/c the expression input has a history of being used to relate a process to something else.

One potential issue may be that in order specify the domain as device, the class device would have to added to RO. Otherwise, we could specify the domain as material entity (I suppose).

[bpeters42]

- I understand the concern about 'input', and I like 'device utilizes material' a lot. - There has been an effort to add 'device' to COB (which I believe is the planned way forward rather than to add classes to RO). This would be generalizing OBI:device to include things with functions beyond scientific investigations, such as 'bottle' etc. @ddooley has been putting some work into that, but I couldn't find good documentation beyond: OBOFoundry/COB#190 .

…

On Thu, Jan 13, 2022 at 2:58 PM Bill Duncan ***@***.***> wrote: Thanks for you input @bpeters42 <https://github.com/bpeters42>! I'm not opposed to changing the label (that is easy to do). But, I am opposed to changing it to capable-of-has-input (or something similar). I think will make the term too difficult for non-ontologists. I understand your point about 'utilizes' being very general. We can be make it little more refined with labels like: - device utilizes - device utilizes material - device uses (material) - device processes (material) I would prefer not to use a label like device utilizes input b/c the expression input has a history of being used to relate a process to something else. One potential issue may be that in order specify the domain as device, the class device would have to added to RO. Otherwise, we could specify the domain as material entity (I suppose). — Reply to this email directly, view it on GitHub <#1480 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADJX2IXFZO2G7KJJ4DCDSQDUV5KJ5ANCNFSM5LUZGRHQ> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Bjoern Peters Professor La Jolla Institute for Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

[ddooley]

An informal call on Friday Jan 21 discussed this and came up with some observations and questions. Of the different kinds of input to a process it seems that some are said to be said to be used by or utilized by devices. E.g. gas and people and oil are all inputs to 'driving a vehicle' process, each a different category of input. 'vehicle' is a device in the process, and some of the inputs have roles with respect to the vehicle.

• I've sent an email off to some contacts at Industrial Ontology Foundry about what shortcut relations they might be using to attach types of process input by devices.
• It occurs to me that questions about whether an input is a consumable by a device in service of process objective, as opposed to material being transformed as a process objective, becomes clearer with process component analysis. For example, both gas and oil are consumed by the vehicle engine. Gas is an input in the subprocess to generate kinetic energy; oil is consumed by (specified input of) an essential lubrication process that occurs in the engine block. Both of these inputs can surface to a general 'driving a vehicle' process (involving archaic combustion engine!).

[wdduncan]

@ddooley These are good questions. I worry a bit as to whether they can ever be answered.
One thought I have is to modify the property chain for utilizes from

capable-of o has-input

to

capable-of o has-primary-input

has primary input is defined in RO as:

p has primary input c if (a) p has input c and (b) the goal of process is to modify, consume, or transform c.

My thought is that using has primary input may alleviate some concerns about arbitrary inputs. If RO included the has_specified_input property, then that could have been used to.

[linikujp]

@wdduncan any follow-up with Industrial Ontology Foundry about vehicle?

[wdduncan]

@linikujp I don't have any contact with the IOF. I thought @ddooley did.

[linikujp]

Hi OBI-ers, will you revive the "utilizes" relation? if you don't, we will use it in CIDO paper. However, I wish you will revive.

[linikujp]

@ddooley Did you follow-up with IOF?

[ddooley]

I did send out a question but didn't hear back. I've just sent a reminder. But regardless, this should be a hopefully shorter topic of discussion at next RO meeting?

[wdduncan]

I have changed the label from utilizes to device utilizes material. See this PR.

[wdduncan]

I don't seem to have permissions to close this issue. Can someone else close it?
cc @ddooley @linikujp @bpeters42

[ddooley]

Closed!