POC Conf. Call 9-29-10

POC meeting, Webex Conference Call; Date: Sept 29th, 2010 10am (PDT)

In attendance: POC members: Laurel Cooper (OSU), Ramona Walls (NYBG), Pankaj Jaiswal (OSU), Alejandra Gandolfo, (Cornell University), Barry Smith (University at Buffalo, NY), Dennis Stevenson (NYBG), Justin Elser (OSU), Justin Preece (OSU).

Absent: Chris Mungall (Lawrence Berkeley National Lab)

Collaborators: none

Acceptance of the minutes from the 9-22-10 meeting? There were no additions, deletions or changes to the minutes of the last meeting.

Status and Update of Progress: PO Release

Steps to get the release out - by the end of the week

-The revised ontology files from the dev browser have been combined with the updated annotation files on the Beta browser for review. Please look it over for any problems- esp with annotations. Still having the issue with the browsing feature and graphical view not working. RW will post item on AmiGO SF tracker

-Updating the annotation files: TAIRs are done (see comments below),

Naama Menda submitted files from SGN for the anatomy ontology (not on beta browser), PJ: The SGN files should be separate files for the different species. Laurel has contacted her and they get it corrected

Gramene- Updates have been made to the files on the SVN.

-updating the AmiGO browser? Justin E is working on it, but it will take a couple of weeks to do it correctly.

-If everything looks good, we will ask Justin E to load them on the live browser

-We will send out announcements to the mail lists and user groups.

-Update info on the release notes page This page will be archived and a new version will be there- with a link to the summary of changes page near the top.

• Swollen eye:

-Has been made a synonym of 'subterranean tuber axillary vegetative bud' after discussing this with Naama. This is a term that the breeders use and swollen eyes may or may not develop into shoots.

Here is the scale that the breeders use to rate the buds/sprouts of the tubers:

1 none

2 swollen eyes

3 sprouts ca. 1/4"

4 sprouts ca. 1/2"

5 sprouts ca. 1"

6 sprouts ca. 2"

7 sprouts ca. 3"

8 new top growth

9 chain tubers

Therefore, 'swollen eye' should be a narrow synonym of 'subterranean tuber axillary vegetative bud' (PO:0025042) (eye)

On a related note (from DWS), using the word sprout could be problematic, since a sprout can be used in many contexts. We already refer to the synonym of 'subterranean tuber axillary shoot' as 'tuber sprout'.

Barry suggested we put this scale in a comment or a link to it (and make singular)

Comments from other reviewers

-Alejandra or Dennis will follow up with Gar Rothwell

Waiting for response from Gar Rothwell and Peter Linder, will copy responses to po-discuss.

-Dennis will check whether there are any other outstanding reviews to come in yet from the folks he contacted. Chris Hardy?

These will have to go into the next set of revisions, if we get anything from them at this point.

Obsoleting terms in the Ontology

For our future work, we need to know: Is it better to use replaced_by or to merge terms.

What are the advantages and disadvantages of each?

Merging (from OBO-Edit User guide)

When term A is merged into term B, the following things happen:

• A's id is assigned as an alternate id of B
• All of A's synonyms are copied to B
• All of A's dbxrefs are copied to B
• A & B's comments are merged
• A & B's definitions are merged
• A's name is assigned as an exact synonym of B.
• All of A's parents become parents of B
• All of A's children become children* of B (Note there is a typo in the user guide here)
• All of A's parent relationships are deleted
• All of A's child relationships are deleted
• A is destroyed

Laurel posted a note on the OE bug tracker regarding these two errors.

What happens to the annotations when terms are merged?

• One advantage of merging is that the comments from the obsoleted term are merged into the comments of the replacement term. But a big disadvantage is that the term is destroyed.

Destroying terms:(from OBO-Edit User guide)

"It is dangerous to use the Destroy command. If a term is destroyed, its id goes back into circulation for use in new term creation, and that id will probably be reused immediately. If the term that was destroyed was ever included in a publicly available version of your ontology, external database and applications that use your ontology may become confused by this id reuse. For example, tools like obodiff will think that instead of destroying a term and creating a new term somewhere else, you have kept the old term in the ontology but changed all its parents, children, its name, its definition, and all its other attributes."

"Applications assume that a term id will be used for a single term and never again. That is why terms are obsoleted instead of destroyed: we need to keep an audit trail of ids that have been used. If it is possible to reuse an id through the use of the destroy command or ontology filtering, this assumption does not hold, and applications everywhere become deeply confused."

Are these terms really 'destroyed' when they are merged? NO, since the ID becomes an alternate ID, and that ID will not go back into use. The alternate IDs show up in the browser as synonyms, so you can find them in the search. Also the definition and synonyms become part of the term that was merged into, so they don't just go away, as in destroying.

Replaced_by: (from OBO-Edit User guide)

• "If an obsolete term is perfectly supplanted by another term, the new term may be specified as a replacement term. Note that a replacement term should only be specified if the obsolete term can be replaced by exactly this term in every case. If this criterion is not met, use a "consider" term instead."

"Replacement terms should be specified when possible, because replacement terms can be used by software agents to do automatic corrections in databases when a term becomes obsolete."

• Also, it is possible to edit the definition and comments of an obsolete term, whereas the destroyed terms are gone forever (but merging does not really destroy the term).

• According to the GO wiki (http://wiki.geneontology.org/index.php/Curator_Guide:_Merge_Split_Move), merging should be used when the two terms have the same meaning, that is, when they would be exact synonyms of each other. That is not the case for the PO terms that were obsoleted during this round of revisions:

-seedling, gametophyte, and sporophyte are narrow synonyms of whole plant

-leaf whorl is a narrow synonym of collective leaf structure

thus, these terms should probably not be merged.

Note: merging should only be used when two terms have the same meaning, which will generally only occur by mistake (a new term was added that has the same meaning of an existing term, because curators were not aware of the existing term) or when two ontologies are merged.

We discussed this at some length and all agreed we would have to evaluate each obsoleted term on a case by case basis. We need to make sure the obsoleted terms are in as synonyms for now (gametophyte, sporophyte, dermal tissue, seedling, leaf whorl) so they show up in searches.

At the ObeEdit user group call on 9-27, CM said we should think carefully about using replace_by to replace a term from the structure ontology with one from the temporal anatomy, since replaced_by implies the same meaning. We felt it was still appropriate in the case of sporophyte, gametophyte and seedling, because even if they temporal term has a somewhat different meaning, it is the case that any annotations from sporophyte (etc.) should now be associated with sporophytic phase (etc.).

• Setting the AmiGO browser to display terms that have been obsoleted.

Laurel posted a message on the AmiGO SF tracker and has been discussing this possibility with Seth Carbon at the Berkeley Bioinformatics group.

In summary, it appears there may* be a variable in the config.pl file that can be set before installation or reinstallstion.

$ENV{AMIGO_OBSOLETE_BEHAVIOUR}

The options are: 'ignore','include_commented', and 'include_all'.

However, Seth was not sure how much of this is in the current version of AmiGO. Chris- Perhaps you know better??

This may be an option in the new version that JE is working to implement.

BS: should allow user to choose by filters if the obsoletes are going to be displayed. Default should be to not display them. Laurel will request on SF AmiGO feature request tracker.

PJ: OBOEdit should have a field that would show date of term creation and date made obsolete, a history with the reason for making it obsolete (this should probably go into the obsolete term's comment field). RW: will request on SF OE feature request tracker.

Updating and Fixing the Annotations

Obsoleted terms:

We need to formalize the procedures in place for making updates to the annotation files on the SVN and in our ontologies.

PJ: Each of the database groups should be responsible for submitting their annotation files to the SVN. User names would preferably not be acronyms- should be recognizable.

Current status:

• TAIR

'po_anatomy_gene_arabidopsis_tair.assoc'

'po_temporal_gene_arabidopsis_tair.assoc'

Laurel obtained up-to-date files from TAIR as the existing ones were 18 months old and committed them to the SVN. Note that the numbers of annotations has changed in this new version.

The changes were made in the files and they were committed back to the SVN. TB sent us a link to a script to run on the files to check them for errors. JE will run the check on them?

Summarized from Email correspondence with Tanya Bernadini: They would prefer us to set up an script to automatically retrieve the files from their FTP site (there was apparently some agreement to do this in the past). They are updated weekly- do we want to reload the files between releases to keep the annotations more up-to-date?

PJ: TAIR should be responsible for submitting their own files to us. LC will correspond with Tanya Bernadini about it.

• MaizeGDB-

Summarized from Email correspondence with Mary Schaeffer: "We prefer to supply them when there are major updates, such as phenotype-stock associations to add for genetic stocks, etc. In general, we are not doing a lot of hand curation. The next major update will probably be between Feb and June, 2011, when the new set of gene models is released, along with their PO associations."

• Gramene- Generally will send them in 2x per year, when they release the new versions. They were not updated since version #30 as there have been no new annotations added. Who at Gramene is going to do this?

• Solanaceae Genomics Network (SGN)-

Naama Menda has obtained SVN access rights (thanks Justin E) and she will look after submitting up-dated files directly to our SVN repository when they are available. She is working on updating them before the release. In order to do so, she will load the po-dev version of po_anatomy_test.obo (from the SVN, version #936) onto their database. That way they can access the tuber terms that we have added for them in the past few months. She is aware that this is still our dev version and said it is not a big issue for them to update their database once it goes on the live site.

• NASC- not active at this moment. File on SVN is 4 years old

• ABRC

Updates to the PO Web pages

List of Priorities:

Documentation page

• We discussed moving the supplemental information on the PO web page to the PO wiki. That way, when we update, we will only have to do it on one site.

This has been done with the POC meetings page, and the Outreach page but several of the other pages are duplicated on the Po-wiki page (for example: Developers style guide, Principles, rationals and key organizing rules of the PO and Ontology accessions.

The 'Principles, rationals and key organizing rules of the PO" should be archive as "version #1" and a new one written on the wiki.

• Annotation file formats- should link to GO page for GAF 2.0?

-did not discuss as we were running out of time

• Introduction to the project slideshow tutorial is really out of date.

-did not discuss as we were running out of time

Reviewers and Acknowledgments Page needs to be updated for current release. -Should be archived as "version #1" and a new one written on the wiki.

Collaborators page needs updated, also called "Acknowledgements" -did not discuss as we were running out of time

Download Page: The OBO and OWL files, are available on the download page from all three sites.

• Are the flat files are being retired?

Tabled: Will deal with after the release is done

• Need a page to list publications that cite the PO- all agreed

-RW created a page on the wiki a few months ago PO Citations, but it has not been populated yet. We already have a Related Publications page -- need to go through this and see if some of those links belong on the Citations page.

The PO Home Page was updated in August 2010 to reflect the new consortium members (OSU, NYBG, Cornell). Do people want links there to their personal sites? Send it to Laurel if you have one you want to use.

We need an "About Us" page similar to Gramene's

Also the link on the PO page "Contact the Plant Ontology Consortium" opens an email to po-dev (which has 78 people on it). Shouldn't this go to po-internal?

Should we have a link to the outreach page from the home page?? Yes

Ramona and Laurel have been gradually updating the documents and other web pages. Will continue to work on this, depending upon priorities..

Other Items

• POC meeting to be held at NYBG in fall 2010. Date TBD. Are we still planning to do this??

Should we invite outside experts, scientists, students and postdocs? Time to do so is getting short.

Please fill out Doodle Poll re. dates that you are or are not available: http://www.doodle.com/6n4pmknkeebk9khg

We need to pick a date for this meeting ASAP so folks who are flying in can get tickets

From 9-15 meeting: Pankaj will check his calender to see if this is possible. Dennis will look at Doodle poll for availability in November.

Meeting will be all day on Friday, November 5th, and morning of Saturday, Nov. 6th. Dennis reserved the NYBG apartment for the nights of Nov. 4th and 5th (may want to reserve it for the 6th as well). Lol may arrive early on the 4th.

• We have been contacted by Stefan Rensing, University of Freiburg, Germany re: setting up an ontology for bryophytes with a focus on the model moss Physcomitrella patens.

Ramona and Dennis responded to him with an explanation of why we chose the current structure. He responded:

'We will aim to setup bryophyte terms with regard to the new structure and (if possible) suggest terms that are broad enough to cover all land plants in the process.'

'FYI, the current, but now obsolete, PO-MO term mapping is at Moss_Ontology'

This site has a list of moss tissue and cell types, as well as life cycle/developmental stages that we could use a a guide for adding new terms.

Issues arising from this:

This groups is setting up their own ontology for their immediate needs, but as the PO includes more moss terms, it will probably become redundant. Should we work with this group to meet their needs quickly, so they don't have to develop their own ontology? They may be able to develop a moss PO slim.

We will keep in contact with this group. DWS: we also have a moss expert at the NYBG who can consult on this (Bill Buck)

*Question from Daniel Lang <daniel.lang@biologie.uni-freiburg.de> (to Pankaj Sept 2009)

"I am looking for estimates of the number of cell types and the number of different tissue types in the gametophytic and sporophytic generations of the following angiosperms:

Arabidopsis lyrata, Arabidopsis thaliana, Carica papaya, Glycine max, Medicago truncatula, Populus trichocarpa, Ricinus communis, Vitis vinifera, Oryza sativa, Sorghum bicolor and Zea mays (current estimate: total 100 cell types)

I've already tried to extract part of this data using the plant_structure associations from the database dump, but I am unsure whether the annotations are comprehensive enough to give a reliable answer?

PO term associations including transitive indirect annotations: Arabidopsis thaliana gametophyte 10 Arabidopsis thaliana sporophyte 229 Oryza sativa gametophyte 4 Oryza sativa sporophyte 145 Zea mays gametophyte 2 Zea mays sporophyte 42

I would be excited to hear your estimates/data on this! Do you know of any web resources, literature or databases that would help us answer these questions?

To conserve the data for future use, I will post the numbers to BioNumbers (http://bionumbers.hms.harvard.edu)."

Has anyone responded to him??

Dennis and Ramona will discuss this during the week and report back at the next meeting

• Setting up conference call with Bruce Kirchoff to discuss the Plant Ontology and the issues he raised at the Botany Meeting

His comments: "I hope you understand that my comments were not directed at you, or the work that you are doing with the plant ontology. I do understand the need for the ontology, at least from a database perspective, and I think that the improvements that you and the plant ontology group have made to the initial ontology are important, and substantial. However, I feel that there are still serious problems with any attempt to bring all aspects of plant morphology together under a unifying ontology. Perhaps the word "ontology" was coined for these controlled vocabularies with tongue-in-cheek, but the attempts to extend the ontology beyond their use in databases tends to remove the jest, at least in my opinion. I very much believe in open discussion as a way of advancing science. It was for this reason that I convinced the Botanical Society to institute the discussion sections that now are a standard part of the annual program. I would be interested in continuing the discussion of plant trait ontology at the next botanical society meetings, in a discussion session. Perhaps you and your colleagues would be interested in organizing one of these sessions with me."

This was scheduled to be done after Sept 20th, so we should probably decide when we want to do it.

Laurel can set up a Doodle poll?? Who wants to take part?

Will schedule this after the release. Maybe invite to meeting at NYBG

PJ suggested preparing a slide (for presentations like IBC or BSA) that shows how PO can be used with data matrices, to appeal to systematists who are not necessarily using genomics. DSW suggested we work with people like David Spooner (USDA Wisconsin) or Sandy Knapp (British Museum) from the Solanacea network, or Paula Rudall (Harvard, already a reviewer) who are already moving between systematics and genomics.

• Status of BSA announcement for bulletin

Notice came out in this quarter's issue of the Plant Science Bulletin. Can be viewed online at: BSA Bulletin on p. 102 (great location - on the first page).

We can also post an online announcement with the BSA when the final release is ready.

• Mailing lists

We need to define the function of each of the mailing lists: po, po-dev, po-announce, po-internal. I think po-announce, po-internal are fine, but what are the other 2 supposed to be for?

In mid-August the new list po-discuss@plantontology was set up and the members of po-internal were added to it. This is the address that will be used in the letters to the reviewers.

We all agreed that these need to be reviewed and we need to deal with the spam problem as well. May be able to use images instead of textual links to email addresses.

We will need to update the links to this lists on the PO web site (http://plantontology.org/mailarchs/mail_list.html)

Tabled for next meeting

• New curator hired at NYBG: Angelica Cibrian has been hired and will be a split appointment between Genomics of Seed Plants project (3/4 time) and the PO (1/4). She will spend part of her time NYBG and rest at NYU. She has been working on using GO terms and annotations to analyze the Seed Plant data, and so can act as an interface between the Seed Plant project and the PO, perhaps provide annotations to the PO.

She will start on Friday

• Fall exhibition at NYBG- DWS: ask PJ about the details of this?

From the grant:

(2) Dr. Stevenson, a co-PI from NY Botanical Garden (NYBG), will involve PO project in the ongoing outreach activities of the NYBG in developing Plant Biology tools for 6-12 teachers in NY City public schools as well as the general public and education programs at NYBG for K-12. NYBG being a Museum gives us an opportunity to reach out to parents and students alike and in ways that a university setting like Cornell may not provide. He will organize three annual exhibitions (one per year) at NYBG.

Deliverables: Years 2-4: Student workshop at NYBG.

Tabled for next meeting

• Plans for a publication to detail the updates in the PO

-This was discussed a bit in May/June, but we have not made any detailed steps towards getting it done yet

-We have been keeping track of the changes on the 'Summary of Changes' page

-Target journal: Current Opinions in Plant Biology or Plant Physiology

-What are peoples feelings about doing this right now? Is it justified at this point?

Tabled for next meeting

*Update from Alejandra about images for the PO

Ramona will go to Ithaca on Oct. 20th to work with Alejandra on loading images onto Plantsystematics.org and linking them with the PO.

PO slims

Tabled for next meeting

GO includes reduced versions of their ontologies (Go slims) for some user groups. The slims are developed and maintained by the users, but publicly accessible as part of the GO flat file. We have one user group (TraitNet) that is interested in creating a PO slim.

We should decide if we want to include that in the PO flat file, and if we want to have any other policies regarding slims.

For example, GO also maintains a generic slim. Should we make something similar (especially as the PO gets more detailed and adds more taxon specific terms)?

Looks like this can be done in OE see: Creating Your Own GO Slim in OBO-Edit. This actually describes how to create slims in an older version of OE. It is similar to creating subsets in the current OE. It does not have the capability to check if the subset will function independently, as needed for a slim.

Berkeley BOP has an online tool for creating slims (http://berkeleybop.org/galaxy/) but the server has been unavailable for the past week and a half.

Priorities for the Next Round of Revisions

Tabled for next meeting

Target date for the next release?

-Merging the plant structure and plant developmental stages ontologies

-Complete compliance with OBO foundry principles

-Adding Musa terms requested by Rosemary Shrestha

-Root terms- submitted by Rich Zobel, adventitious roots

-Legume terms submitted by Austin Mast

-New terms requested by reviewers (from Ahrestani and Kramer)

-is_a parents for remaining terms (will require work on upper level structure for non-material entities)

-add terms for non-angiosperm structures

-better organization of descendants of portion of plant tissue

-convert to intersection_of relations?

-Convert existing definitions to genus-differentia form?

-work on PGDSO

-tuber growth and development stages, open on SF since 6/2009

-add links to more images through PlantSystematics.org

PJ: Lol and Ramona will focus more on adding annotations

-Others??-

Upcoming meetings 2010-2011:

• Latin American Congress of Botany

LACB. La Serena, Chile, October 4-10;

No one from our group will be able to attend.

• Infectious Disease Ontology Workshop

Dallas, Dec. 8th and 9th, 2010

Organized by Lindsay Cowall (Duke University) and others, under the auspices of the National Center for Biomedical Ontology. IDO Workshop December 2010. Note that relations between the IDO and other ontologies is one of their provisional goals.

Pankaj will represent the PO with regard to plant infectious diseases. This is especially relevant to the PO as infectious disease agents (viruses and bacteria) are widely used in plant genetic research.

For additional information on the IDO see http://www.infectiousdiseaseontology.org/Home.html

• Plant and Animal Genome XIX Conference

January 15-19, 2011 PAG 2011

Registration is open, abstract deadline Nov 1, 2010. To qualify for reduced registration rates, must register by October 15th.

Lol and Ramona should plan on attending and meeting with collaborators. Set up a computer demonstration, and possibly a Phenote workshop. Perhaps Dennis and Alejandra could also attend, and we could all meet there. PJ is not attending as he is teaching.

* International Botanical Congress (IBC2011)

July 23rd-30th 2011, Melbourne, Australia

Registration is open- deadline for abstracts: 31 October 2010, Important dates

Dennis and Alejandra are planning to attend IBC2011 and speak in other symposia.

Symposium proposal was accepted, 'Bio-Ontologies for the Plant Sciences' under the genetics, genomics and bioinformatics theme.

Ramona contacted the organizers about whether or not it is possible to get a room with Internet for outreach activities, but hasn't gotten a response yet. Is this something we still want to pursue?

Pankaj, Ramona, and Angelica Cibrian (Harvard and ANMH, Genomics of Seed Plants) are scheduled to give talks and must submit abstracts by October 31st. Three additional speakers will be chosen from abstracts that are submitted.

Pankaj will give the introductory talk on general use of ontologies, GO, genomics, etc, and Ramona will present the talk on the Plant Ontology.