Represent exposure

[pbuttigieg]

Represent the various forms of exposure noted in the SDGs. Following on from #6 and #18

[pbuttigieg]

From @cmungall's input https://github.com/SDG-InterfaceOntology/sdgio/issues/18#issuecomment-142820660:

What does 'exposure' mean here?

I can think of different senses

direct exposure to X: the subject comes into contact with X by overlapping the same spatiotemporal space at some point (e.g. is burned by the fire) indirect exposure to X: the subject does not overlap but is in the vicinity (e.g. subject smells the smoke, and needs to abandon their house) risk of {direct,indirect} exposure to X (e.g. subject has increased insecurity through living in fire-prone area exacerbated by lack of access to fire protection services) exposure in the acturial or business sense (I won't attempt an elucidation, out of my league)

An experimental biologist may think in terms of (1). This is the sense used in terms of the chemical-toxicogenomics exposure ontology (https://github.com/CTDbase/exposure-ontology-draft). I suspect these sense required here is more along the lines of (3) or possibly incorporating aspects of (4)

Here is an existing UN definition of exposure:

https://www.wmo.int/pages/prog/drr/resourceDrrDefinitions_en.html Exposure is the total value of elements at-risk. It is expressed as the number of human lives, and value of the properties, that can potentially be affected by hazards. Exposure is a function of the geographic location of the elements [Source: UNDP (2004): Reducing Disaster Risk: a challenge for development. A global report (M. Pelling, A. Maskrey, P. Ruiz, L. Hall, eds.). John S. Swift Co., USA, 146 pp,]

[pbuttigieg]

Perhaps pato-ontology/pato/issues/74 is of relevance. We could probably express 'x is exposed to y' through processes, however.

[pbuttigieg]

@cmungall EXO looks very relevant. If its domain could be broadened to allow cases from 1-3 (not sure about 4 either) it would be a good general purpose exposure ontology. Even if not (in which case, should it not be called the toxicology exposure ontology or similar?) , cases like (1) will be relevant to SDGIO too (exposure to radiation).

We could and should list it here.

[mark-jensen]

I added ExO to the list of relevant ontologies. Shall exposure always be associated with risk and vulnerability? The quality of being exposed seems neutral wrt to harm, but the exposure process could be directed that way. This seems to be what ExO is doing with stressors and receptors.

[pbuttigieg]

I would think the receptor (in SDGIO, generally an individual, human population, or human community) has a set of innate vulnerabilities (~ dispositions to be harmed) which an exposure event may or may not trigger, depending on whether a) the entity which a receptor is exposed to can act as a stressor to that entity and b) the receptor possesses any defences which mitigate or nullify the influence of the stressor.

As for risk, I believe @phismith had some thoughts on this. I would say we would need to differentiate between the risk of being exposed to some event and the risk of realising a vulnerability associated with that exposure. Risk seems like some sort of probabilistic quantification on a disposition of a bearer to participate in some process.

[pbuttigieg]

@mark-jensen We are moving to add vulnerability, resilience, and robustness into PATO. This will let us express entities around exposure more accurately. See https://github.com/pato-ontology/pato/issues/67

@cmungall @phismith Given @cmungall's cases, could we use the sense of RO's interacts with and generalise to the degree that exposure is very close to interaction? This is more geared to capture (1) direct exposure and (2) indirect exposure. When one thing is exposed to another, they are both causally impacted, we just choose to focus on one (for a variety of reasons). exposure/interaction: a process in which the processes executed by one entity are causally connected to those executed by another. i.e.when X is exposed to Y, both the processes in which X and Y are agents in change due to the exposure.

When X and Y are on roughly the same sort of thing, this seems workable: A human individual exposed to a population of pathogens is quite similar to these two entities interacting (~ interaction with host. The same can be said for a human population and a forest ecosystem they are settling. The forest is 'exposed to' humans.

If X and Y are different sorts of entities, things are less clear, e.g., exposure of a human individual (material entity) to an explosive volcanic eruption (process). The interaction here concerns the participants in the process of eruption that the human encounters. There is an interaction between the human and some portion of the pyroclasts, tephra, lava, etc produced by the volcano or the particulates left over long after the volcano has stopped erupting. The processes these entities undergo on contact with the human (cooling, hydration, etc) complement those (are causally connected with those) the (parts of) the human undergoes (undergo) (heating, dehydration, etc).

Thoughts?

As a footnote, GEMET's exposure seems limited to exposure to some form of electromagnetic radiation.

[phismith]

On Thu, Sep 24, 2015 at 7:39 AM, Pier Luigi Buttigieg < notifications@github.com> wrote:

I would think the receptor (in SDGIO, generally an individual, human population, or human community) has a set of innate vulnerabilities (~ dispositions to be harmed) which an exposure event may or may not trigger, depending on whether a) the entity which a receptor is exposed to can act as a stressor to that entity and b) the receptor possesses any defences which mitigate or nullify the influence of the stressor.

As for risk, I believe @phismith https://github.com/phismith had some thoughts on this. I would say we would need to differentiate between the risk of being exposed to some event and the risk of realising a vulnerability associated with that exposure

​This sort of matter is dealt with quite well here http://ontology.buffalo.edu/ido/Dispositions_and_IDO.pdf, I believe.​

​ See especially under blocking dispositions BS​

.

— Reply to this email directly or view it on GitHub https://github.com/SDG-InterfaceOntology/sdgio/issues/21#issuecomment-142905008 .