environmentontology / environmental-exposure-ontology Goto Github PK
View Code? Open in Web Editor NEWModular environmental exposures ontology
License: Other
Modular environmental exposures ontology
License: Other
The interacts_with_an_exposure stressor_via object property (http://purl.obolibrary.org/obo/exo.obo#interacts_with_an_exposure_stressor_via
) needs to have a definition.
The dbGAP Multi-Ethnic Study of Atherosclerosis and Air Pollution (MESA Air) captures
longitudinal data in support of investigating cardiovascular disease outcomes in response to air
pollution exposure. The study includes air pollutant, emission and exposure variables that are
well suited for semantic modeling via the Environmental and Exposure Ontology (ECTO).
has stressor o has input subpropertyof: has stressor
So
has stressor some smoking and has input some nicotine
will classify under
has stressor some nictoine
2019-09-20 18:52:56,509 ERROR org.obolibrary.robot.ReasonOperation - Equivalence: http://purl.obolibrary.org/obo/ENVO_3100011 == http://purl.obolibrary.org/obo/ENVO_09200021
2019-09-20 18:52:56,510 ERROR org.obolibrary.robot.ReasonOperation - Equivalence: http://purl.obolibrary.org/obo/ENVO_01001483 == http://purl.obolibrary.org/obo/ENVO_00010504
They seem to be equivalent.. Why?
TERIS - Teratogen Information System
database designed to assist physicians or other healthcare professionals in assessing the risks of possible teratogenic exposures in pregnant women.
http://depts.washington.edu/terisdb/index.html
see http://depts.washington.edu/terisdb/samples/Betamethasone.htm
for sample record
REPROTOX https://reprotox.org/
Contains summaries on the effects of medications, chemicals, infections, and physical agents on pregnancy, reproduction, and development.
see sample entries here: https://reprotox.org/samples
MotherToBaby
Service of the non-profit Organization of Teratology Information Specialists, is dedicated to providing evidence-based information to mothers, health care professionals, and the general public about medications and other exposures during pregnancy and while breastfeeding.
https://mothertobaby.org/
@mellybelly who should be the contact? @nicolevasilevsky?
I think GO needs to start using this and migrating exposure annotations away from GO, see geneontology/go-ontology#15848 (comment)
Questionnaires often have Qs like "how often do you engage in vigorous activity"
vigorous is defined something like: lifting heavy objects, intense physical labor
Answers are grouped by daily, once a week etc.
We should import a physical activity ontology here.
cc @mateolan
There are number of developmental stage terms (e.g., blastula stage, embryo stage etc.). These should be grouped under a comment parent term, such as 'life cycle stage' (http://purl.obolibrary.org/obo/UBERON_0000105
)
There is an extra space in 'abdominal radiography related exposure'
New term: exposure to personal care or consumer product chemicals.
Parent: exposure to chemical with application.
Definition: A exposure event involving the interaction of an exposure receptor to a personal care or consumer product chemical. Exposure may be through a variety of means, including through the air or surrounding medium, or through ingestion.
This comes from the CDC Report on Human Exposure to Environmental Chemicals (https://www.cdc.gov/exposurereport/pdf/FourthReport_UpdatedTables_Volume2_Mar2018.pdf). This is part of a class at Oregon State University. cc: @putmantime
Wikidata has various classes of use under hazard
In particular, chemical hazards link to exposure routes and phenotypes, e.g. strychnine exposure - this seems to come from CSC NIOSH
try this query
SELECT ?cause ?chebiId ?causeLabel ?exposure ?exposureLabel ?symptom ?symptomLabel ?route ?routeLabel WHERE {
?exposure p:P636 ?statement .
?statement pq:P780 ?symptom .
?statement ps:P636 ?route .
?exposure wdt:P828 ?cause .
?cause wdt:P683 ?chebiId .
SERVICE wikibase:label {
bd:serviceParam wikibase:language "en" .
}
}
This query would eventually be integrated into dipper (see monarch-initiative/dipper#363 for discussion on wikidata statement model), but for now in exploratory phase it's useful to see what data WD has info on exposures on.
Inferred parents in MRE:
Are these all correct?
One way to manage this is to say that if someone is the recipient of violence then they have been exposure to violence constructed using violent behavior in NBO. To say that someone is exposed to violence in their surroundings, we could construct a "violent environment" class in ENVO. Thoughts?
Related to #6
ECTO's classification is based entirely on CHEBI and our OWL defs. MRE is manual.
In have aligned ECTO and MRE, and compared classifications:
I will break down the categories here:
More to be added later
Considering exposures to mother/fetus dyads here and trying to identify some route options we could use to describe fetal exposures:
Fetus is exposed to chemical (ex. Cocaine, exposure direct to fetus)
Amniotic fluid route (combo of oral and transdermal exposure to fetus)
Placenta route
-Maternal/fetal exchange of nutrients, water, gases, and waste substances
-Significant mediator to fetal chemical exposure, including environmental, supplement/drug exposure to mother, food based exposures etc.
Fetal exposure secondary to maternal exposure
Example:
Fetus is exposed to uterine artery vasoconstriction d/t exposure to chemical (cocaine)
Fetus is exposed to blunt trauma during motor vehicle accident
Maternal/fetal route, Maternal route, ??
-Can be used to describe instances in which the impact of a chemical/ exposure is indirectly impacting the fetus, i.e. mediated by the mother
-I think this can be used for instances like the motor vehicle accident as well? But we will need to consider an appropriate “route” option for MVAs and other types of trauma items when talking about a free living human since we don’t really have a good option for that yet.
Just a head's up that the SDGIO is dealing with "exposure" as defined by UN bodies. This is more along the lines of "assets exposed to risk". We had a crack at "exposure" as a process in the more general sense too.
'exposure process'=def. "A process which is causally within or downstream of another process whose direct or indirect effects cause one or more entities to realise one or more of their dispositions.
+ comment: "Consider that when "E died of exposure to P", E died as an outcome of the processes which were initiated by E's participation in P, in which E's dispositions to react to P (or its parts or participants) were realised."
+ editor note: "This class may be replaced by a class in EXO, pending a clarification on the scope of its meaning. Right now it is essentially the same as 'interaction'. https://github.com/SDG-InterfaceOntology/sdgio/issues/21"
ECTO needs to contain classes that will be useful for others modeling immune responses. The latest release has a pattern for exposure to organism. I would like to keep this pattern for other use cases, but this is likely not sufficient for the immunity use case. For example, someone could be exposed to a cat without being exposed to toxoplasmodium. Someone could be exposed to toxoplasmodium without developing an infection. Then someone could be exposed to an infected person or be infected themselves. I could use some help thinking through this to arrive at a suite of patterns that satisfies use cases.
@bpeters42
src/Makefile:
make: *** No rule to make target ecto-edit.obo', needed by
ecto-core.owl'. Stop.
src/resources/wikidata/Makefile:
make: *** No rule to make target hazard-all.obo', needed by
combined.obo'. Stop.
Any idea why I have the errors? Thanks.
What's the difference?
As you can see, Nico and I are starting to migrate ECTO to odk and adding patterns that are a bit different from what was there originally. We are trying to be in alignment with PECO and ExO. There is still a lot of work to do. Feel free to take a look at the patterns in the dosdp-patterns directory (the others are just in reserve and may not be used). I would love some feedback.
Analysis of all MRE terms that can NOT be easily factored into DPs or mapped to ECTO.
I suggest we focus purely on the chemical set for now. Many of these are due to missing CHEBI IDs. The others are probably some trivial systematic thing.
I think this may be a simple gap in the DPs
I think this may be a simple gap in the DPs
I think this may be a simple gap in the DPs
Note - I couldn't find terms from just one ontology, and I couldn't find an ontology term for infrasound.
Lower priority - should really have this in separate ontology
Exposure through residues left on surfaces etc
I am unable to open ecto-edit.obo in Protégé, I get an error that says: Level: ERROR Time: 1515454262730 Message: An error occurred whilst loading the ontology at Could not load imported ontology: Cause: /Users/vasilevs/git/environmental-exposure-ontology/src/ontology/imports/ncbitaxon_import.owl (No such file or directory). Cause: {}. Any thoughts on how to troubleshoot?
inferred superclasses for exposure to ethylparaben
:
are these correct?
they all come from CHEBI:
Our role pattern is 'has exposure stimulus' some ('chemical entity' and 'has role' some %s)
, so if CHEBI says that C has role R, then an exposure to C will be classified as an exposure "to" R. Is this right in all cases? Let's look.
I think we need to distinguish between
'has exposure stimulus' some ('chemical entity' and 'has role' some %s)
'has exposure stimulus' some ('chemical entity' and 'has role' some (%s and realized-by some process))
E.g.
Should we always name these 2 classes, to avoid confusion, and for maximum precision? Also a 3rd class for when we know the role is not realized?
If we follow this pattern for 'hormone', we would have exposures to chemicals with potential hormone roles, and exposures to chemicals with realized hormone roles. The former is what most people would query by. For example, we would expect BPA to come back in a query for xenestrogen, whether or not it actually realized this role in any given exposure. Of course it's hard to determine whether in any given exposure the role is realized, so the existence of the subclass may seem odd.
It may be simplest in the short term to continue to use the uncommitted version (1), but for some cases, specifically those roles involving some kind of human intention, we make more specific subclasses.
A number of definitions appear to be generated programmatically. In these cases, it is helpful to add an annotation to say which script and/or design pattern generated the definition.
Hi! Combing through the UDN survey info, and below are some of the exposures indicated within the survey that we do not currently have modeled in ECTO. Will continue to update this ticket.
Chemicals/Mixtures: (most all substances currently included from chebi terms)
Will require consideration for a new env. material class (mixture):
Other stuff:
Requires ENVO terminology:
Indoor environment:
Irregular environment exposures: ** an odd mix of quite specific items, rec for more universal terms
exposure to environmental copper (living near copper mine), using copper atom and will adjust route as necessary
exposure to chromium, route water (chrome 6 hexavalent chromium in water supply) (using chromium atom)
exposure to iron, route water (high iron in well water)
exposure to electric substation (living near electric substation, included power plants of various energy sources)
exposure to pesticide
insecticides,
rodenticides,
herbicides,
fertilizers,
fungicides (includes
ethyl dibromide/1,2-dibromoethane
exposure to PFOS, route water (using perfluorooctane-1-sulfonic acid)
exposure to PFOC, route water (in water near home. also in fish population from rivers) did they actually mean PFOAs?? (adding perfluorooctanoic acid)
exposure to extreme temperatures (did add extreme heat, may need to req extreme cold)
exposure to extreme weather (a variety of additions provided)
exposure to extreme noise
exposure to methyl tert butyl ether (gasoline additive in well water)
exposure to well water (community or private) (have included underground water)
exposure to city/town water (drinking, showering, bathing)
exposure to filtered water/water filtering system (water(processed) from FOODON?)
exposure to bottled water (look for FOOD ON term)
exposure to rain water or cistern
exposure to river, lake, or pond water
exposure to consuming holy water (consider FOODON term?)
exposure to fumes (general or nondescript air content, also from fires, and from living near methamphetamine lab)
exposure to radon atom
exposure to farm animals/cattle, or pet animals (dogs, cats, small furry animals?)
exposure to water damage at the home (leaks, broken pipes, flooding)
exposure to fresh water source (swimming, not consumption, we have pool options here, what is the goal of this term?)
exposure to an animal/vector bite (cat, dog, monkey, insect, tick, other)
exposure to environmental allergens
Occupational/hobby exposures:
Lifestyle exposures:
I kind of understand (I think) what you are trying to get at with the term environmental condition. However, every term under it is a process (surgical process to be exact). This makes me wonder if what you really need a relation between exposure events and what you are calling 'exposure conditions'. Currently, what you are calling 'conditions' are really processes, and this can cause ambiguities down the line.
from Bjoern:
Many of exposures, in particular experimental ones such as administration
of a compound to an animal model, changes in bedding, etc. are covered in
OBI or in OBI scope. There are also classes in OBI for unplanned exposures
not really in scope, but we needed to include them as a placeholder to keep
definitions consistent and there wasn't always a natural home for them
(e.g. travel to an endemic area of Malaria; living in the same household as
an active TB case, naturally occuring infection). Randi presented that work
at ICBO, and we are actively looking to coordinate these exposure terms
across ontologies.
This is great, I think many of these are complementary to what we may have.
So at a very minimum, I would beg you to include OBI in the list of
ontologies you are including in your considerations. Better yet, I would
ask you to coordinate this work with us, minimally at the higher level.
(administered exposure / naturally occurring exposure / inferred exposure).
Our plan in ECTO was to be neutral about intentionality. Biologically an exposure to X is an exposure to X. ECTO is already somewhat ridiculously compositional, and to have {admininstered,natural} exposure to {C} via {Route}
could get too unwieldy.
Was this what you had in mind?
Wikidata has a lot of exposures associated with symptoms/disease. From NIOSH?
https://github.com/cmungall/environmental-conditions/tree/master/src/resources/wikidata
See geneontology/go-ontology#12523
The desired hierarchy might look like:
The basic pattern would be {change,increase,decrease} {nutrition,calory}. We would get elemental roles or defined classes for the latter from the food ontology.
I think we should avoid overloading is_a and treat starvation/malnutrition as disjoint conditions with caused_by axioms linking to the reduction hierarchy (or the increased hierarchy, for overnutrition)
Currently these classes appear to be doing the same kind of work. Why is there a distinction? They seem to be the same thing.
Compiled list of medications indicated in UDN survey, includes both classes and individual medications:
- [ ] exposure to sedatives
Will focus on inclusion of the chemical components only, may include generic or brand name within synonyms
- [ ] exposure to tranquilizers or anti-anxiety drugs
- [ ] exposure to painkillers
- [ ] exposure to stimulants
- [ ] exposure to marijuana, grass or hash (smoke)
- [ ] exposure to cocaine
- [ ] exposure to hallucinogens
- [ ] exposure to inhalants or solvents
- [ ] exposure to heroine
OTHERS NOTED
Per the EPR survey, responses are categorized as either listing currently used medications or classifying medications intended to treat certain systems:
Cardiovascular health:
Endocrine Health
Mental Health
Respiratory Health
Other Conditions
Reproductive Health
@cboerkoel - please provide any additional information I may have missed.
UDN survey supplement free responses, may need to include in ECTO:
- [ ] exposure to vitamin supplements
NOTE: providing exposure to vitamins as chemicals from CHEBI, will include supplemental and food based forms.
- [ ] exposure to mineral supplements
- [ ] exposure to phytochemical supplements
I'm looking for a term for pm2.5 exposure. I found 'exposure to fine respirable suspended particulate matter' ECTO:0000113
. It seems like this could be a match depending on how strictly "fine" is interpreted. Is this right?
E.g. nlm cde browser aggregates multiple CDEs
exposure CDEs:
https://cde.nlm.nih.gov/cde/search?q=exposure&page=1
When we say exposure to {caesium, lead, mercury, copper, ...} do we mean
X molecular entity
etc
it seems safest to opt for the most general (X molecular entity
), and if we need more specific subforms, add more specific classes.
some examples:
is_a CHEBI:33676 ! d-block molecular entity
is_a CHEBI:33673 ! zinc group molecular entity
is_a CHEBI:25196 ! mercury molecular entity ***
is_a CHEBI:25706 ! organomercury compound
is_a CHEBI:22648 ! arylmercury compound
is_a CHEBI:25193 ! mercuribenzoate
is_a CHEBI:25194 ! mercuribenzoic acid
is_a CHEBI:33255 ! alkylmercury compound
is_a CHEBI:39152 ! mercury difulminate
is_a CHEBI:59445 ! chlormerodrin
is_a CHEBI:35113 ! elemental mercury
is_a CHEBI:16170 ! mercury(0)
is_a CHEBI:25197 ! mercury cation
is_a CHEBI:36561 ! mercury coordination entity
is_a CHEBI:28216 ! methylmercury chloride
is_a CHEBI:31823 ! mercury dichloride
is_a CHEBI:33050 ! dimercury dichloride
is_a CHEBI:33210 ! dimercury diacetate
is_a CHEBI:33211 ! mercury diacetate
is_a CHEBI:36569 ! triiodomercurate(1-)
is_a CHEBI:36573 ! mercury dicyanide
is_a CHEBI:36574 ! diamminemercury(2+)
is_a CHEBI:36576 ! tetraiodomercurate(2-)
is_a CHEBI:36577 ! tetrakis(thiocyanato)mercurate(2-)
is_a CHEBI:49639 ! mercury dibromide
is_a CHEBI:49659 ! mercury diiodide
is_a CHEBI:49727 ! (acetyloxy)mercury(1+)
is_a CHEBI:33674 ! s-block molecular entity
is_a CHEBI:33296 ! alkali metal molecular entity
is_a CHEBI:37128 ! caesium molecular entity ***
is_a CHEBI:33988 ! caesium hydroxide
is_a CHEBI:37130 ! caeside
is_a CHEBI:60270 ! caesium ion
is_a CHEBI:49547 ! caesium(1+)
is_a CHEBI:63039 ! caesium chloride
may not be sufficient to classify radioactive isotopes
ECTO Class Label | Exposure to ultrafine respirable suspended particulate matter via inhalation |
---|---|
ECTO URI | ECTO:0000977 |
Axiomatic Definition of ECTO class | exposure event and (has exposure route some inhalation) and (has exposure stimulus some ultrafine respirable suspended particulate matter and (part of some air)) |
Classes and Properties from other ontologies used to form the axiomatic definition | exposure event = ExO:0000002 has exposure route = RO:0002242 has exposure stimulus = RO:0002309 ultrafine respirable suspended particulate matter = ENVO:01000416 part of = RO:0000050 air = ENVO:00002005 |
The occurrent branch of `ecto.owl (http://purl.obolibrary.org/obo/ecto/releases/2019-09-20/ecto.owl) does not include BFO:process.
consumed items that are not foods from UDN survey:
- [ ] exposure to consuming non-food item(s)
Food items (FOODON)
Dietary behavior (EPR) (FOOD ON, can model specifics of these if we like?)
frequency of eating at: fast food restaurants, sit down restaurants, buffet restaurants, takeout restaurants, grocery stores (prepared food), cafeterias, vending machines, on-street vendors, other (gas station, quick mart, bakery)
consuming snacks
meal timing
Pgs. 20-25 EPR B are focused on FFQ type data. What do we want to do?
This was approved during the obo operations committee today. Please create the appropriate YAML files in https://github.com/OBOFoundry/purl.obolibrary.org.
ALPHA-GLUCOSIDASE INHIBITORS | |||
---|---|---|---|
T-TYPE CALCIUM-CHANNEL BLOCKER | |||
SLOW-RELEASE DIHYDROPYRIDINES OTHER THAN NIFEDIPINE OR AMLOPIDINE (L-type calcium channel blocker | |||
Erectile dysfunction drugs | Unsure, this would fall under PDE5 inhibitors or some category? | ||
K-CHANNEL BLOCKERS | |||
Weight loss drugs | |||
antiarrythmic classes 1A, 1B, 1C, and 3. |
at the moment we are only adding Sulfonylurea Antidiabetic Agent and not distinguishing 1st and 2nd generation sulfonylureas.
@matentzn
This wikidata query:
PREFIX wd: <http://www.wikidata.org/entity/>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wdt_cause: <http://www.wikidata.org/prop/direct/P828>
PREFIX wikibase: <http://wikiba.se/ontology#>
PREFIX p: <http://www.wikidata.org/prop/>
PREFIX p_route_of_administration: <http://www.wikidata.org/prop/P636>
PREFIX ps: <http://www.wikidata.org/prop/statement/>
PREFIX ps_route: <http://www.wikidata.org/prop/statement/P636>
PREFIX pq: <http://www.wikidata.org/prop/qualifier/>
PREFIX pq_symptoms: <http://www.wikidata.org/prop/qualifier/P780>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX bd: <http://www.bigdata.com/rdf#>
PREFIX RO: <http://purl.obolibrary.org/obo/RO_>
PREFIX upstream_positive: <http://purl.obolibrary.org/obo/RO_0002304>
PREFIX has_exposure_route: <http://purl.obolibrary.org/obo/RO_0002242>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
CONSTRUCT {
?cause rdfs:label ?causeLabel .
?exposure rdfs:label ?exposureLabel .
?symptom rdfs:label ?symptomLabel .
?route rdfs:label ?routeLabel .
?statement
rdfs:label ?genLabel ;
rdfs:subClassOf ?exposure ;
rdfs:subClassOf [
a owl:Restriction ;
owl:onProperty has_exposure_route: ;
owl:someValuesFrom ?route
] ;
rdfs:subClassOf [
a owl:Restriction ;
owl:onProperty upstream_positive: ;
owl:someValuesFrom ?symptom
]
}
WHERE {
?exposure p_route_of_administration: ?statement .
?statement pq_symptoms: ?symptom .
?statement ps_route: ?route
bind(concat(?exposureLabel, " via ", ?routeLabel) AS ?genLabel)
SERVICE wikibase:label {
bd:serviceParam wikibase:language "en" .
}
}
doesn't succeed in binding genLabel - I assume because label service is called after the fact.
Replacing the service call with explicit rdfs:label clauses slows the query down too much
proposed new MRE links based on ECTO
In this case the mapping from MRE to ECTO is probably wrong.
ECTO:
MRE:
MRE:0000173 ! Exposure to inorganic compound *** [DEF: "Being exposed to a compound that is not organic, i.e.,that does not contain carbon."]
Here compound is likely used in the sense of two or more elements. The ECTO/CHEBI usage is more general, encompassing elements.
There is no CHEBI class for 'inorganic compound'
https://en.wikipedia.org/wiki/Inorganic_compound has a whole discussion on different usages
To align with BFO / OBO core, exposure event should be a subtype of BFO:process (see #57).
In ecto.owl (http://purl.obolibrary.org/obo/ecto/releases/2019-09-20/ecto.owl)
, exposure event is a top level term.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.