I'm not sure how we're going to handle taxonomy in EnvO. I suppose mapping to a 
taxonomy is an option. I think we should have general terms like "tree" and 
"shrub", however. Thoughts?

We can MIREOT a subset of NCBITaxon quite easily.

But this may not correspond exactly to what we need. Sometimes the 
morphological grouping may not be monophyletic. There is no shrub or tree in 
NCBITaxon.

Also, when we say "pine tree" in ENVO we don't necessarily mean any instance of 
an organism of taxon pinus. We mean to exclude the forms before it's fully 
grown - e.g. a seed can be an instance of pinus.

Given this I think we should mint new ENVO IDs where we need them, Have text 
defs of the form

Tree:
"A tree is plant that has an elongated stem, or trunk, which supports the 
photosynthetic leaves or branches at some distance above the ground"

Pine tree"
"A pine tree is a tree of taxon pinus (NCBITaxon:nnnn)"

The logical definition depends on your ontological commitment w.r.t taxonomy, 
this is often controversial, but we made a decision with the OBO library 
NCBITaxon ontology:
http://www.obofoundry.org/wiki/index.php/NCBITaxon:Main_Page#Organisms

So formally
'pine tree' EquivalentTo envo:tree AND NCBITaxon:pinus

tree is too hard to define (N+S) in OWL, but we could have necessary conditions:
tree SubClassOf 
   starts_after some PO:mature stage??
   has_part some PO:bark??
etc

Another way to do this is to treat treeness as a quality, and say
'pine tree' EquivalentTo ncbitaxon:pinus and has_quality treelike

I did try to import a subset of NCBI Taxon to a local version of EnvO to see 
how it worked but my computer kept crashing. Since I wasn't able to find any 
example of a tree class in other ontologies I started defining 'pine tree' and 
similar terms without specifying if they were trees or not. So I created some 
additional definitions to fit all the terms we had hoping to find a solution 
for 'tree' later:

DEFINITION      terrestrial feature=def. a prominent or distinctive aspect, 
quality, or characteristic of a terrestrial biome.
IS_A        environmental feature


DEFINITION  land plant=def. a plant that is a terrestrial feature.
IS_A        terrestrial feature
SYNONYMS    embriophyta

PREF NAME   pine 
DEFINITION  pine=def. a pinaceae that has biennial, rarely triennial, cones with 
each year's scale growth distinct, forming an umbo on each scale. Cone scale 
base broad, concealing the seeds fully from abaxial view. Seed without resin 
vesicles. Seed wing holding the seed in a pair of claws. Leaves with primary 
stomatal bands adaxial (above the xylem) or equally on both surfaces. 
IS_A        pinaceae
SYNONYMS    pinus [NCBITaxon:3337]

So both your solutions are very elegant, specially treating tree as a quality 
but I don't know what the experts would find more useful. 

Is there a way you prefer me to submit future similar terms?

Thanks,
Julian

There are two aspects here. Probably the sample which was collected from the 
Pine tree was from a 'portion of' not the whole tree. Secondly, the sample 
(probably another organism) found on the host pine tree was collected. In such 
cases we are looking at the pine tree as an ecological environment subject. 
This is equivalent of human body that hosts a plethora of other microorganisms 
too. 

What's the rationale for putting 'human/animal/plant body' in the envo.

Also this seems like a XP (cross product) term 

PO:0000003 whole plant 
intersects_with NCBI_tax:3312 Coniferophyta

Since the term 'pine' is too generic I suggest using Coniferophyta instead of 
the one suggested in the discussion ' Pinaceae'

Ref: PO
http://plantontology.org/amigo/go.cgi?view=details&search_constraint=terms&depth
=0&query=PO:0000003

Ref NACI_Tax
http://www.ncbi.nlm.nih.gov/Taxonomy/Browser/wwwtax.cgi?mode=Undef&id=3312&lvl=3
&lin=f&keep=1&srchmode=1&unlock

Pankaj: you're assuming the term is requested for recording samples. This isn't 
necessarily the case. I assumed the term is required for describing a 
particular biome or habitat.

> PO:0000003 whole plant 
> intersects_with NCBI_tax:3312 Coniferophyta

See my original comments. This is actually equivalent to just saying 
NCBITaxon_3312. We need to (presumably) exclude seeds etc, assuming the goal is 
to define the grown tree. If that isn't the goal, and the goal is to have a 
term that describes any instance of Coniferophyta, regardless of stage, then 
just use NCBITaxon.

Juilian: try OntoFox - http://ontofox.hegroup.org/

Chris, I am saying the same too. A tree as an ecosystem/habitat/biome.

There are different issues here.
First what is a "tree" (of course it is a whole plant) but by definition a tree 
is a "woody perennial plant having a single main stem that may be short but is 
usually considerably elongated, has generally few or no branches on the lower 
part and is crowned with a head of branches and foliage.

But as Pankaj said: it is also an ecosystem/habitat/biome. It just depends on 
the request.

The synonym Pine tree: "A pine tree is a tree of taxon pinus (NCBITaxon:nnnn)" 
Actually Pinus is a genus, if you say taxon it can be any taxonomic group of 
any rank, such as a species, family, or class.

Chris,  I am fine with using only with the taxon id as long as it is meant for 
saying that this species/taxon (tree) is part of a habitat. However the context 
changes (as suggested by me) when we look at it from the host/ecosystem 
perspective.

Ale: your definition of tree is fine, but one can use the same to define woody 
perennial shrubs as well. Remember in a Banyan/Fig tree sometimes the main 
trunk is gone/dies in older stages. So it won't hold. I suggest leaving it as a 
synonym of Whole plant.

Hi all,

I am sure I am missing much of the nuances. I just wanted to clarify that as 
Chris mentioned the term was required to describe a particular habitat.

Regarding Pinaceae vs Coniferophyta, in the corpus that we parsed there are  
conifer forests (or coniferous forests) and pine forests. So even though pine 
is too general I guess the researchers were making a distinction.

And thanks for the onto fox tip!

Best,
Julian