POC Conf. Call 3-29-11

POC meeting, Webex Conference Call; Date: Tuesday Mar 29th, 2011 10am (PDT)

In attendance:

POC members: Laurel Cooper (OSU), Ramona Walls (NYBG), Justin Preece (OSU), Justing Elser (OSU), Pankaj Jaiswal (OSU), Marie Alejandra Gandolfo; (Cornell University, Dennis Stevenson (NYBG)

Absent: Barry Smith (University at Buffalo, NY), Chris Mungall (Lawrence Berkeley National Lab)

Collaborators: none

Acceptance of the minutes from the POC_Conf._Call_3-22-11? There were no changes, additions, or deletions.

Issues with the ontology arising during writing:

This section is for questions about the PO that came up while were writing the manuscript. These are issues that need to be addressed by the curators, preferably before publication and the next release.

Rename Plant Structure Ontology (PSO) to Plant Anatomy Ontology (PAO)?

Reason: better reflects the new upper structure, highlights the changes we have done over the past year

PAO is consistent with other anatomical entities in OBO foundry, like HAO.

We agreed to change the name to Plant Anatomy Ontology. Since this name or PAO does not actually occur in the ontology anywhere, and it is just a name for one branch of the PO, changing it is not a problem. Also, the file on the SVN, where the separate versions of the files are stored, the PAO branch file is already called plant_anatomy.obo, so this name is more consistent.

We should check with CM to see if we should change the namespace, which is currently plant_structure to plant_anatomy, and what the ramifications of such a change would be.

This name Plant Structure ontology is used on the PO webpage and in other spots where the individual files are linked to (such as Bioportal)- so it should also change in those places. Will need to be changed on the AmiGO Browser

Use of ontology id's in definitions

We have been inconsistent in where we put ID's for terms from other ontologies that we reference in the PO.

For the top level PSO terms, we have CARO ID's in the definitions, but also have them as xrefs.

For plant cell, we reference GO ID in the comment, but also have it as an xref.

For the top level PAO terms that reference CARO, we should remove the ID from the definition and add it as a definition dbxref. We should put the definition and ID of the CARO term into the comment, like we have done with GO for plant cell.

As a general guideline, whenever we use a definition from another ontology for a term, but alter it to fit PO, we should put the other ontology ID plus POC:curators into the definition dbxrefs. This gives credit to the source of the definition

Note: We should check that the browser will automatically create the hyperlinks, or we can also add them there. We should request CARO to the GO Dbx database, along with SourceForge and PATO.

Does CARO/PATO need to have a website in order to do so? We could use the link to Bioportal

LC: I added a statement and example explaining this to our PO_Developers_Guide.

embryonic plant structure

-Do we want dual parentage for embryonic structures? Should we use intersection_of relations, instead of asserting dual parentage?

-Will the intersection_of relations be displayed by AmiGO? No they are not

-Here is a figure of how it would work, if the is_a relation to the "normal" plant structure were asserted, and the relation to embryonic plant structure were inferred using intersection_of relations, using embryonic apical cell as an example:

pi represents participates_in relation

dotted lines are relations that are inferred by the reasoner.

• This method defined embryonic plant cell as is_a: embryonic plant structure, intersection_of: is_a plant structure, intersection_of: part_of embryo.

Because the intersection_of relations specify sufficient conditions, any plant cell that has a part_of relation to embryo will be classified as an embryonic plant cell whenever the reasoner is on. Similar relations could be used for other embryonic plant structures.

• This allows us to classify cells as plant cells, tissues as plant tissues, etc., but still lets them have a relationship to embryonic plant structure, while other structures, like embryo proper, could remain as is_a children of embryonic plant structure.

This is the method specified as the best_practice for OBO foundry (at least orally, if not actually in their principles).

We agreed that it was okay to start using intersection_of relations, and to use them to classify embryonic plant structures.

For each new live release, we will create a version of the plant_ontology file that asserts the relations implied by the reasoner. For the time being, we should do the assertions manually (either in OboEdit, or with a script, which we manually check), to be sure that they are relations we want. This would also normally be done regularly while editing, by using the reasoner.

Do we want to release two versions -- one with implied links asserted and one without -- or should we keep the file without only for curators?

in vitro plant structure

What are the boundaries for in vitro structure?

What about a whole plant or parts thereof that are grown in culture? Suppose someone had an annotation for a leaf from an in vitro grown plant. Where should they put it?

Also, we talked about using derived_from relations for in vitro structures (e.g., cultured cell derived_from plant cell). Do we still want to do that? How to assure that the derives from relations always hold?

We discussed whether or not we want the class for in vitro plant structures to only include plant structures that have not yet fully differentiated into tissues or organs (or whole plants, for that matter). We can add a subclass of cultured callus for "differentiating callus" (if it can be defined correctly).

It was suggested on the current definition could to be changed to reflect the fact that it should only include non-differentiated structures.

LC: This makes no sense at all- where do you draw the line?- we still have not defined the boundaries. I am reopening the SF tracker. Whether or not something is in vitro has nothing to do with whether or not it is differentiated.

Perhaps in vitro structures should be children of their respective whole plant parts (cell, tissue, and embryo). Just like an embryonic plant cell is still a plant cell, an in vitro plant cell is still a plant cell.

Would it be possible to use intersection_of terms to classify structures as in vitro, so they could also be children of their respective whole plant parts, like we did for embryonic plant structures? Would have to have some discriminating character, like "participates_in in vitro growth," which is not currently part of the PO. Might be more practical at this point to use dual parentage.

We will work further on this and discuss this more next week

has_part relation

For more details- please see the POC Technical Issues Page

How will we deal with has_part relation? for this release? for future releases?

We confirmed that the reciprocal part_of and has_part relations were what was actually cause the problem with loading the obo file onto our browser:

-JE tried to load the dev browser with the current plant_ontology.obo file with only one reciprocal part_of/has_part relation, but no other known problems, and it would not load.

-Then he loaded the dev browsers with a file that had only has_part relations but no reciprocals, and it worked fine.

-Still have to test loading it onto beta browser with annotations

Report from JE @ tests and annotations etc:

PJ created some smaller ontology files with has_part relations to test in the beta browser. JE confirmed that they load fine, which means that the has_part relation does not interfere with loading. We still have the issue of how annotations pass through the has_part relations. PJ and JE are working on it with CM.

For the time being, the annotations can be manually assigned to the correct terms, if necessary

For the time being, we can continue to use has_part relations, but we will have to revisit how to deal with them before the next release.

participates_in relation

Need to get these in. Should we make a test file to see if these are going to cause issues?

RW- I have been putting them in the developer's file. Don't seem to interfere with loading. Show up on the dev browser, but need an icon.

Question: If a parent term (e.g., plant gametangium) has a participates_in relation, do we need to specify participates_in relation for all its children (e.g., archegonium and antheridium), or will the reasoner know to apply the relation to the children?

Should not need it for children if it is transitive:

A is_a B, and B participates_in C, then A participates_in C

and

A part_of B, and B participates_in C, then A participates_in C

Answer (from Chris Mungall): participation isn't transitive. (in OboEdit by default)

what you want to say is that a is part of b, and b participates in c, then a participates in c

unfortunately you can't do this in OE yet - use a text editor and add the following line to the declaration for participates_in:

[Typedef]

id: participates_in

holds_over_chain: part_of participates_in

- this gives the reasoner the information that an apical cell participates_in gametophyte phase via the part_of relation

(really this axiom should be in RO, and there should be some mechanism where you can just import it but things are a bit primitive right now)

RW: I did this and it worked.

LC: Should we request this feature on the OBO-Edit tracker?

It is okay to continue to add the participates_in relations to the developers' file. PJ prepared an icon using the letters P-i in green, which we can put on the browser.

Issues arising from last week's meeting:

New definitions for gametangium, micro- and megagametophytes, and sperm and egg cells were accepted. The SF tracker items will be set to pending, so anyone can comment on them for the next few weeks.

Recap of Gametangium, antheridium, archegonium and related terms

Proposed new definitions from last week's meeting:

plant gametangium (PO:0025124): A plant organ that produces one or more gametes that are located in it and is part of a whole plant in the gametophytic phase.

participates_in gametophytic phase, only_in_taxon bryophytes plus pteridophytes plus gymnosperms (=never_in_taxon angiosperms)

Comment: The embryo sac and pollen replace the function of gametangia of producing gametes in gymnosperms (for antheridia) and angiosperms (for both antheridia and archegonia).

-Should not say disjoint_from megagametophyte, because we will define megagametophyte such that archegonium is a part of it (see below).

*Use of "located in" rather than "contained in."

Contained_in is used when an organism has a cavity or space that contains some material object that is not part of it (such as a fetus in a womb).

Located_in is used when one class located in another is also part of the class it is located in.

- We will use located_in for the same reason a human mother's egg is part of the mother rather than contained in the mother. Also, we can put pictures on Plantsystematics.org that show gametangium development, and show how the egg cell derives from one of the same population of cells as the rest of the gametangium.

• New subsets were added for bryophytes, pteridophytes, and gymnosperms, in addition to the subset for angiosperms. Will put terms in proper subset(s) until we can use the only_in taxon relation.

• Added new relation participates_in.

This definition of gametangium works for algae. Could add a child term oogonium. Do we want to add a subset for algae? At this point, we don't have any terms specific to algae, but several that apply.

antheridium (PO:0025125): A plant gametangium that produces antheridial sperm cells that are located in it.

LC: note on the dev browser it says: "produces antheridium sperm cells"

Comment: A antheridium has a single outer layer of non-sperm producing cells called the sterile jacket layer and sporagenous cells on the interior. There are no antheridia in seed plants.

only_in_taxon bryophytes plus pteridophytes (or never_in taxon seed plants), add relation sterile jacket layer part_of antheridium

Should add antheridial initial as a synonym of generative cell.

archegonium(PO:0025126): A plant gametangium that produces an archegonial egg cell that is located in it.

Comment: There are no archegonia in angiosperms or Gnetum or Weltwischia.

only_in_taxon bryophytes plus pteridophytes plus gymnosperms (never_in taxon angiosperms, Gnetum and Weltwischia)

Note: there are no examples of unicellular archegonia in the green algae or land plants, so we don't need to worry about it

New terms and proposed definitions from last week's meeting

We should obsolete PO:0020092 megagametophyte and PO:0020091 microgametophyte, and create new terms that can encompass dioicous bryophyte gametophytes as well as the gametophytes in angiosperms. Embryo sac and pollen grain will be children of these terms, as well as archegonial megagatephyte and antheridial microgametophyte.

LC: for us non bryophyte experts dioicous and monoicous

megagametophyte(PO:0025279): A whole plant in the gametophytic phase that produces only egg cells.

Comment: Megagametophytes in bryophytes only produce egg cells, but do not develop from megaspores, because there is no heterospory in bryophytes. In some pteridophytes, megagametophytes develop from megaspores. In angiosperms, Gnetum and Weltwischia the megagametophyte is greatly reduced. The megagametophyte in angiosperms is an embryo sac.

microgametophyte (PO:0025280): A whole plant in the gametophytic phase that produces only sperm cells.

Comment: Microgametophytes in bryophytes only produce sperm cells, but do not develop from microspores, because there is no heterospory in bryophytes. In some pteridophytes, microgametophytes develop from microspores. In seed plants the male gametophyte is reduced to a pollen grain.

embryo sac (existing term, PO:0025074): A megagametophyte that is located in an ovule in an ovary.

Comment: Produces an embryo sac egg cell.

develops_from megaspore, only_in_taxon angiosperm

Do we need a separate class for the microgametophytes in Gnetum and Weltwischia? Usually don't use "embryo sac" for these taxa.

For the time being, any annotations for these two genera should just be placed under microgametophyte or megagamethopyte.

pollen: A microgametophyte that is located in a pollen sac.

Comment: Produces pollen sperm cells.

develops_from microspore, only_in_taxon seed plants

-Would pollen grain be a better name for this? Pollen implies all of the pollen grains, collectively.

will add pollen grain as an exact synonym

archegonial megagametophyte: A megagametophyte that has as parts one or more archegonia.

Comment: Megagametophytes in bryophytes only produce egg cells, but do not develop from megaspores, because there is no heterospory in bryophytes. In some pteridophytes, megagametophytes develop from megaspores.

only_in_taxon bryophyte plus pteridophytes

antheridial microgametophyte: A microgametophyte that has as parts one or more antheridia.

comment: Microgametophytes in bryophytes only produce sperm cells, but do not develop from microspores, because there is no heterospory in bryophytes. In some pteridophytes, microgametophytes develop from microspores.

only_in_taxon bryophyte plus pteridophytes

New definitions for egg cells and sperm cells from last week's meeting

archegonial egg cell (PO:0025122): An egg cell that is produced by and located in an archegonium.

only_in_taxon bryophytes plus pteridophytes plus gymnosperms (or never_in taxon angiosperms), part_of archegonium

embryo sac egg cell (PO:0025123): An egg cell that is produced by and located in an embryo sac.

only_in_taxon angiosperm, part_of egg apparatus

For Gnetum and Weltwischia, should just use parent terms (egg cell or sperm cell).

antheridial sperm cell (PO:0025120): A sperm cell that is produced by an antheridium.

only_in_taxon bryophytes plus pteridophytes (or never_in seed plants), part of antheridium

pollen sperm cell (PO:0025121): A sperm cell that is produced by a pollen grain.

Comment: A pollen sperm cell consists of a nucleus without a cell wall.

only_in_taxon seed plants, part_of pollen

New Physcomitrella and related terms

These items were postponed until next week.

Collective plant structures:

vascular and non-vascular shoot systems

The Moss Ontology did not request these terms, but suggest adding them in order to classify shoot systems.

non-vascular shoot system: A shoot system that does not have as part a vascular system (or vascular tissue?).

comment: Does not have any xylem of phloem, but may have other conducting cells, such hydroids or leptoids. Can occur in both the gametophytic and sporophytic phases of non-vascular plants.

subsets for bryophytes and pteridophytes

Could we use the relation "lacks_part"? Then we could define this class using intersection_of relations, which would be better. CL has used this (lack_plasma_membrane_part). Works better in OWL; translates into something like has_part exactly 0 (some ?Y) [need to check this].

Children of non-vascular shoot system will be gametophore (see below) and thallus. If we used the lacks_part relation and intersection_of relations, we could make gameotophore a child of shoot system, and the reasoner would infer it was a non-vascular shoot system.

Can't use participates_in gametophyric phase, because the sporophyte of a bryphyte is also a non-vascular shoot system (since it doesn't have roots or vascular tissue).

vascular shoot system: A shoot system that has as part a vascular system (or vascular tissue?).

comment: Has xylem and/or phloem. Occurs only in the sporophytic phase of vascular plants.

participates_in sporophytic phase, has_part vascular system

If we accept these terms, suggest changing definitions of leaf and non-vascular leaf to "A leaf that is part of a non-vascular/vascular shoot system. This would be better than the current definitions which reference their taxonomy (part of a vascular or non-vascular plant) and would help with reasoning.

root-bourne shoot system (PO:0004544) should be is_a vascular shoot system

gametophore

Definition supplied by Moss Ontology: The leafy moss plant. The gametophore is the adult form of the moss gametophyte and bearer of the sex organs (gametangia). Ref: Reski (1998): Development, genetics and molecular biology of mosses. Botanica Acta 111, 1-15.

Suggest is_a shoot system.

Proposed def.: A non-vascular shoot system that consists of the shoot axes and non-vascular leaves of a plant in the gametophytic phase. participates_in gametophytic phase

Comment: A gametophore is the leafy part of the gametophyte of mosses and leafy liverworts, excluding the protonema. In mosses, gametophores develop from buds that form on the caulonema. Antheridia and archegonia arise on the gametophore.

This will work as long as we can call the shoot system in leafy liverworts a gametophore as well.

gametophore bud

The term "bud" has been requested: Def'n supplied by Moss Ontology: A structure produced by a caulonema and able to develop into a gametophore or a stem that includes an apical cell able to develop into a gametophore. The earliest recognizable stage of gametophore development. Ref: Bill and Nancy Malcolm (2006): Mosses and other Bryophytes, an illustrated glossary, second edition and altered by David Cove

Suggest a new term gametophore bud. This could be a child of bud (PO:0000055: An undeveloped shoot system).

Proposed def'n: A bud that develops into a gametophore.

Comment: Occurs in mosses (and leaf liverworts?). Develops from a caulonema cell in mosses.

participates_in gametophytic phase, has_part shoot axis apical cell

Thallus

From Parihar: A simple vegetative plant body not differentiated into root, stem and leaf, and lacking vascular tissues.

From Schofeld: A flattened gametophore in which no leaf-like organs dominate the structure.

Saying a thallus is a gametophore implies that it is a shoot system and therefore a collective plant structure (CPS). As a CPS, it must have more that one organ. This is not always true for a thallus (sometime they branch, so you could say they have multiple axes, but it is a stretch. Maybe better to classify it as a whole plant.

alternative proposed definitions:

1) A flat gametophore with no distinct organs.

Comment: A thallus is a whole plant in the gametophytic phase in liverworts and pteridophytes and develops from a short-lived protonema. Roughly two dimensional growth results from division of a single apical cell. Although there are no distinct organs, there may be tissue differentiation and dichotomous branching.

or

2) A whole plant in the gametophytic phase that has a flat growth form and no distinct organs.

Comment: A thallus is a gametophyte of liverworts and pteridophytes and develops from a short-lived protonema. Roughly two dimensional growth results from division of a single apical cell. Although there are no distinct organs, there may be tissue differentiation and dichotomous branching.

Add prothallium (for ferns) as a synonym?

Upcoming meetings 2011:

* ICBO 2011 Second International Conference on Biomedical Ontology July 26-30, 2011 Buffalo, New York

ICBO

LC contributed to the workshop proposal "From Fins to Limbs to Leaves: Facilitating anatomy ontology interoperability" Authors: Melissa Haendel, Chris Mungall, Alan Ruttenberg, David Osumi-Sutherland and Laurel Cooper (Accepted) LC is working with the other organizers to develop the workshop plans and a call for papers.

call for papers

Full-Day Workshops Schedule:

July 26 9am-6pm The Ontological Representation of Adverse Events: Working with Multiple Biomedical Ontologies

July 27 8.30am-4pm Facilitating Anatomy Ontology Interoperability

July 26 6.30pm-9pm Evening Workshop: Common Logic

July 27 4pm-8pm Evening Workshop: Doctoral and Post-Doctoral Consortium

- LC will attend and represent the PO, will submit a short paper/poster for the workshop (deadline April 10th).

*Plant Biology 2011, Aug 6-10th, Minneapolis, Minn

Plant Biology 2011

Early-bird registration ends May 13.

Gramene will be putting together a workshop again, focusing on pathways. PJ will present a PO poster.

Abstract deadlines: Your abstract must be submitted by March 11 if you want it to be considered for a minisymposium talk.

For inclusion on the program memory stick and in the program book, abstracts must be submitted by May 27.

TAIR is organizing an Outreach Booth and we are invited to take part.

* International Botanical Congress (IBC2011)

July 23rd-30th 2011, Melbourne, Australia

Registration is open Important dates

Symposium proposal was accepted, 'Bio-Ontologies for the Plant Sciences' under the Genetics, Genomics and Bioinformatics theme.

Dennis, Alejandra, Pankaj and Ramona are planning to attend.

Early bird registration deadline - Extended 1 March 2011 Deadline for registration by presenters 1 March 2011

See IBC 2011 Bio-Ontologies Symposium wiki page for more details