-
Notifications
You must be signed in to change notification settings - Fork 0
ENVO_GOLD
Ongoing ENVO/NMDC effort to create ENVO terms for GOLD/EBI-MGNIFY. Chris Mungall, Pier Buttigieg in support.
Previously I referred to the project as Taxonomy of Ecosystems
Current working version of GOLD-Paths-to-MIxS-ENVO-triad-mapping old.
The following is the old EBI_MGnify_Biome spreadsheet where I am working to create ENVO terms for EBI as well as the JGI GOLD database, from Summer 2019 (now older version).
ENVO editors rolling agenda/minutes
- I notice that in the editors note of the relation part of it says:
only an independent continuant can be part of an independent continuant; only an immaterial entity can be part of an immaterial entity ...
From this are we to understand that an environmental zone, subclass of an immaterial entity can't be part of an environmental system? This is contradicted by the axiom of the class littoral zone: part of some aquatic environment. Is this an inconsistency or is this ok?
-
Can I use the
part ofrelation defined in the robot template intended forenvironmental systemto add an axiom for zones for example in the case of addingpart of somefreshwater littoral zone to the EBI axiom:root > Environmental > Aquatic > Freshwater > Lentic > Littoral zone -
If yes to above, create ENVO zone term for
Limnetic zonesee wiki page -
extend logical gaps in term coverage?
I see various examples of where I could help to fill out the hierarchy by creating semantics for missing cases. For example in the thermal springs hierarchy under the Hot (42-90C) sub-hierarchy we have classes for Acidic, Alkaline Neutral, and Sediment. Under the Tepid (25-34C) sub-hierarchy we only have classes for Sediment not the others. There are many examples of this throughout. Should I go ahead and fill such gaps? If so should I also add the synonym and EBI Biome Path axioms even if they do not yet exist in EBI or GOLD?
- Deep subsurface
could use the site class continental subsurface zone if it's cool to use the relation part of a zone like in issue 2.
- Viriome
SENT TO BONNIE
In EBI there currently exist three Viriome classes root > Environmental > Terrestrial > Soil > Viriome, root > Host-associated > Human > Respiratory system > Pulmonary system > Viriome and root > Host-associated > Mammals > Respiratory system > Pulmonary system > Viriome (no marine or others). You'd think EBI/JGI GOLD would have some systematic way of organizing their omics where you could search just on virome. Unless I'm missing something, as far as I can tell there isn't. Which leads me think that perhaps I could create a bunch of Viriome classes in addition to the regular biomes I'm creating ENVO terms for. Do you know if there is some GSC sanctioned way of differentiating viral from other omics data? In the original Ivanova_et_al-2010-Environmental_Microbiology paper (where this EBI/GOLD classification schema was originally conceived) they talk about working to standardize on community type (e.g. microbial or viral), but I'm unsure if the has actually been done. Perhaps a niche for us to fill, especially with your viral background?
- Arthropoda gut segments
maybe duplicated terms?
root > Host-associated > Arthropoda > Digestive system > Gut > P3 segment
The term must be sourced from the description given in this paper Cellulolytic Systems in Insects (Arthropoda include termites).
Where they describe the five segments of the termite hindgut (from front to back): ileum (P1), enteric valve (P2), paunch (P3), colon (P4, this region is sub-divided into P4a and P4b in some cases), andrectum (P5).
Paunch: an enlarged part of the termite hindgut specifically developed to harbor symbiotic protozoa, bacteria, and archaea.
According to this paper Internal ornamentation of the first proctodeal segment of the digestive tube of Syntermitinae (Isoptera, Termitidae) in a logical extension of whats described in the other paper those 5 segments are refereed to as proctodeal segments which gives us the root > Host-associated > Arthropoda > Digestive system > Gut > Proctodeal segment
However I'm confused that there is also the root > Host-associated > Arthropoda > Digestive system > Hindgut > P3 segment perhaps what I found earlier was the hindgut p3 and there's a different gut p3? or maybe the two terms refer to the same thing and are duplicates? Neither hindgut p3 or gut p3 have any samples in EBI. On JGI gold both classifications hindgut p3 and gut p3 lead to the same samples hence I suspect it's simply term duplication.
- JGI classes not in EBI set
I notice that on the JGI GOLD (see Biosample Ecosystem Classification and Ecosystem Tree) there are some classes which don't exist in EBI, for example in Host-associated, Arthropoda there are terms like Ant dump, Beetle, Mosquito etc which aren't in EBI, however these examples have no associated data. They also have various levels of Unclassified which itself have samples such as Host-associated microbial communities from haemolymph of Banana aphid from Africa - Burundi. I wonder if this is the next phase of the project, after completing what's in EBI's hierarchy to go and make the JGI terms, I'd have to manually crawl through their website but it's doable. It would be a bit more involved to go and make terms for all the unclassified metagenomes.
- Uncharacterized
Perhaps similar to the JGI's Unclassified, EBI has 3 terms for Uncharacterized:
None of which have any associated samples in EBI, however the Mollusca Digestive system:Ceca has an Uncharacterized in JGI GOLD with one sample Bankia setacea gill microbial communities from Puget Sound Washington USA - Caecum Enriched Hence I'm unsure what to do about these Uncharacterized samples.
- NCBITaxon metagenomes hierarchy
the NCBITaxon has a hierarchy of [metagenomes](http://purl.obolibrary.org/obo/NCBITaxon_408169 with subclasses for organismal metagenomes as well as the ecological metagenomes from which I believe data on NCBI is annotated under their Organism field. Perhaps working these terms into the EBI semantics would help to cross over with NCBI better? UniProt also uses these terms. When annotating submitted data to EBI They have annotation with a metagenomic tax ID as an essential step. In the example they show the NCBITaxon_646099 human metagenome (so I know it's that same hierarchy).
- check for duplicates of EBI terms
for example the ceca cecum were duplicates, check root > Host-associated > Mammals > Digestive system > Fecal and root > Host-associated > Mammals > Digestive system > Large intestine > Fecal and root > Host-associated > Mammals > Gastrointestinal tract > Intestine > Fecal for the same.
They appear to be the same term, should collapse this into one term likely more similar example with intestines etc.
- Digestive system vs Gastrointestinal tract
I'm wondering if the Gastrointestinal tract classes should be subclasses of Digestive system classes in mammals. I think yes
- upland, lowland
Maybe need to create classes for upland and lowland to deal with the classification of freshwater rivers wiki page. Perhaps add biome term for upland river as well as a superclass for freshwater rivers generally?
- microbialite
create envo microbialite see wiki or better this definition from here
Microbialites are “organosedimentary deposits that have accreted as a result of a benthic microbial community trapping and binding detrital sediment and/or forming the locus of mineral precipitation” (Burne and Moore, 1987, pp. 241–242).
Maybe this can be a subclass of mass of biological material and maybe move stromatolite mat as subclass and give it a definition (currently missing one). Maybe add sister class thrombolites as well.
- bog
have mire and peatland and types of bogs but the hierarchy isn't clear. see wiki bog page iucn-uk-peatland as well as Soros funded Central European University page on bogs. As well as the wiki mire page
- Diffuse flow
According to this article:
“Diffuse flow” is a catchall term to describe low-temperature (< 0.2° to ~ 100°C) fluids that slowly discharge through sulfide mounds, fractured lava flows, and assemblages of bacterial mats and macrofauna. Diffuse-flow fluids are generally mixtures of cold seawater and either magmatically heated fluids, conductively heated seawater, or both.
Will probably need to add a term for diffuse flow to envo, and link it to EBI Diffuse flow. Maybe as subclass of hydrothermal fluid
- bore hole water
bore hole water has no definition or axioms. Add axiom linking to borehole which is also missing a definition.
- seamount and submarine volcano
ISSUE POSTED
The current envo seamount has exact synonym submarine volcano, upon examination of the wiki submarine volcano page, we see that they are potentially active magma-releasing volcanoes, whereas seamounts are formed from volcanic activity, but (as far as I can tell) aren't active. Should we perhaps change the exact synonym in seamount away from submarine volcano and make a new submarine volcano class?
- athalassic saline lake, epilimnion and hypolimnion
WORKED INTO SEPARATELY POSTED ISSUE
For the athalassic saline lake microbiome as well as for the epilimnion and hypolimnion class, I'm making the assumption that this was intended to be in a lake as opposed to more general non-marine saline water bodies. I'm not sure if this is correct the few existing instances of annotated data are of lakes. In principle there could be a non-lake, non-marine saline water body, which isn't also covered by the Salt crystallizer pond but I guess the chances are low?
- Non-marine Saline and Alkaline > Saline > Thalassic
WORKED INTO SEPARATELY POSTED ISSUE
A bit of a paradox a non-marine system which is of or by the sea. A lagoon is the obvious example, however, they are probably more such of environments and perhaps a term describing such broader set of environments is in order.
- fungus gardens
Ants such as termites and leaf cutter ants (as well as some beetles) cultivate fungal gardens see encyclopedia Britannica page as well as this page by Dr. Scott Turner, and the merriam-webster page.
For termites worker ants will make fungus combs by masticating woody material evacute it (out of their bodies) by semi-digesting it and using the resulting pseudofeces, which is then molded into fungus combs. The fungus combs are innoculated with fungal spores, which grow into hyphae and digest the woody cellulose material. A termite colony will amasses a large number of fungus combs, which they will gather into a series of galleries (fungal galleries) atop the nest called a fungus garden. Amazing.
Original article about it Fungus-Growing Ants
- gulf
Perhaps we can make an ENVO gulf class subclass of coastal inlet see wiki page here
Fracking water
Make hydraulic fracturing wastewater as subclass of industrial wastewater http://purl.obolibrary.org/obo/ENVO_01000964
- Biogas plant
can use biogas and industrial building in axioms.
- Wet fermentation
subclass of Biogas plant, where it does a wet fermentation process (could make dry fermentation too?)
- Continuous culture
make chemostat class, subclass of bioreactor
- inoculum
may or may not be necessary for these bioreactor classes.
- Bioremediation
may EBI/JGI Bioremediation classes, could maybe make a bioremediation environment or similar class ideally as subclass of anthropogenic environment perhaps also a bioremediation process class to link it to the environment class.
-
Biotransformation
-
Microbial enhanced oil recovery
see wiki page
- Microbial solubilization of coal
see this paper
- Solid waste
we have waste material which could serve as superclass.
- Composting
could make subclass of soild waster compost waste, we have compost to serve as axiom.
- grass
we don't have the environmental material grass.
Also why is solid environmental material is a subclass of ice?
- Landfill leachate
have leachate to serve as superclass.
- Mine water
either use mine drainage or make class for mine wastewater subclass of industrial wastewater maybe add axiom in mine drainage to this.
- Petrochemical
make class such as petrochemical wastewater subclass of industrial wastewater and add axiom to petroleum
- Anammox
perhaps make ECOCORE class for anammox organism subclass of chemolithoautotroph I believe.
- Nutrient removal
these all are wastewater treatment bioreactors
- Biological phosphorus removal
make a class such as phosphorus removal bioreactor
- Dissolved organics (aerobic)
I think a term for dissolved organic carbon (like I've created in PMO to push to ENVO) should work here hopefully. same for Dissolved organics (anaerobic) I'll add the aerobic and anaerobic qualities.
we'll need terms for the following, which can perhaps be subclasses of the biotic mesoscopic physical object as sister classes to tissue culture?:
-
Lab enrichment
-
Defined media
-
Aerobic media
-
Anaerobic media
-
Marine media
-
Undefined media
-
Lab Synthesis
is a mess with data from Human blood serum, Synthetic, and Deep sea subsurface microbial communities... Don't know what to do with this.
We also need to figure out what to do with Modeled, Simulated communities (DNA mixture), Simulated communities (microbial mixture), Simulated communities (sequence read mixture), Sanger. Could use the ENVO mock community term. Which is a subclass of cell culture which is incorrectly a subclass of tissue culture confusion between eukaryotic cell cultures and prokaryotic enrichment cultures.
Link to https://github.com/EnvironmentOntology/envo/issues/613.
- Chlorinated (drinking water)
Could maybe use drinking water (purified), however that's less specific (could be any type of treatment). Should we perhaps create a foodon class for chlorinated water, or perhaps a pato quality for chemically treated with subclass chlorinated?
- aquaculture
Is it ok to use FOODON aquaculture as it's a subclass of geographic places and regions (subclass of material entity)?
- epiphyte
maybe add to ECOCORE? They have endophyte.
There exists epiphyte http://purl.obolibrary.org/obo/FLOPO_0900030 in FLOPO but I'm not convinced about this ontology.
- green algae
There is no obvious class for green algaea unlike brown it's not just one phylogenetic clade. According to the wiki page:
The green algae (singular: green alga) are a large, informal grouping of algae consisting of the Chlorophyta and Charophyta/Streptophyta, which are now placed in separate divisions, as well as the potentially more basal Mesostigmatophyceae, Chlorokybophyceae and Spirotaenia.
Perhaps create an ECOCORE class?
- symbiosis-related classes
We'll need classes for the following: Ectosymbionts, Extracellular symbionts, Intracellular endosymbionts (including primary and secondary), as well as Epibionts. Infectious Disease Ontology IDO currently has a symbiont class. Could this be used, or could this be imported into ECOCORE and then make subclasses for the others?
- Fish
acording to the wiki page
Fish are a paraphyletic group: that is, any clade containing all fish also contains the tetrapods, which are not fish. For this reason, groups such as the "Class Pisces" seen in older reference works are no longer used in formal classifications.
Hence there is no one NCBItaxon class there are several: including Actinopterygii, Hyperotreti, Hyperoartia etc. It maybe a bit silly but perhaps we could use the FOODON fish as it gives a good definition of how humans commonly refer to fish.
- invertebrate
There is no single clade for invertebrates, will have to find another base term, which is how I stumbled into the NCBITaxon metagenomes hierarchy as they have invertebrate metagenome.
- coral
No single clade to describe group
- Microbial
No single clade to describe group
- Protists and Protozoa
Neither are proper phylogenetic clades, perhaps additions to ECOCORE? have Protozoa in Vaccine Ontology. Protozoa apear to be a subclass of Protists, have a NCI Thesauraus purl for Protista
- reptiles
reptilia are not a monophyletic clade there are three main groups Crocodylia (crocodyles), Lepidosauria (lizards), and Testudines which we think of as reptilia. The highest common ancestor Sauria also includes birds.
- Spiralia
Spiralia are an unranked group including annelids and molluscs. Perhaps This is too broad or I should look at the data examples of what Spiralia microbomes have actually been sequenced.
has no definition. Comment saying term was sourced from another ontology needs to be checked for compatibility with uberon.
in Mammalian Phenotype Ontology not in UBERON maybe ask them to take it in?
- Bacteriome
see wiki page specialized organ insects have to host bacterial symbionts. Make and push UBERON term for this.
- Endoperitrophic space
EBI has it as subclass of stomach but according to this wiki page the endoperitrophic space is within the midgut.
- periodontal pocket
exists as MP term but could it be in UBERON?
-
paunch
-
Proctodeal segment
I wasn't able to find microbiome terms which to link to the following EMPO subset terms as axioms:
non-saline aerosol environment
environment determined by a plant substance
environment associated with an animal part or small animal
environment determined by a biofilm on a fungal surface
environment determined by a biofilm on a non-saline surface
environment determined by a biofilm on a plant surface
environment determined by a biofilm on a saline surface
environment determined by a biofilm on an animal surface
environment determined by a metazoan secretion
non-saline planetary subsurface environment
saline environment Used subclass saline water environment
I noticed that in human there are no Foregut Midgut or Stomach classes Should these be added when I "fill in" un-represented classes?
I had previously added environment determined by a metazoan secretion to animal feces associated microbiome but fecal environment was more appropriate. It would probably be good to fill in all possible uberon classes for all organismal groupings, for example animal excretory system could make the list of all the organismal groupings and cross with the uberon terms (making sure it makes sense and they exist in those animal etc).
add saline lake sediment when I clean up the athalasic/thalaci issue.
First take at the project, classifying ecosystem taxonomy via clustering functional gene abundances from metagenomes see the github rep here