POC Conf. Call 10-4-11

POC meeting, Webex Conference Call; Date: Tuesday Oct 4th, 2011 10am (PDT)

In attendance:

POC members:

Absent:

Collaborators: none

Any changes or corrections(additions/deletions, etc) needed in the minutes from the POC_Conf._Call_9-27-11?

Issues and status reports for the Upcoming Release

For more details of the changes in the upcoming release, see: Summary of Changes to PO September 2011

Timeline of Upcoming Release

After today's edits to palea and lemma, the plant_ontology.obo file will be considered complete for this release, barring any major issues that are noticed.

Once that is done, we send out a notice to our collaborators and ask them to update the annotation files. This involves asking TAIR to suspend their automatic script that runs on a weekly basis, as they have to access the dev version of the ontology file, not the live one. Usually, we give the database groups about a week to update the files on their end and often this entails LC and JE just looks after making the required changes.

We discussed the option of proceeding in a two-stage release where the "finished" plant_ontology.obo file is loaded onto the live browser. We will run the QC checks on annotation files (check for obsoletes and column 16 annotations) and then we send out a notice to the database groups that they need to update their files. Once the new files come in they will be uploaded to the database.

Updating of Dbxrefs

Our version of AmiGO has a built-in set of dbxrefs, in a Perl file, and it is not reading the current GO file.

In order to update these, this file has to be so will have to be edited manually by JE (or JP).

We decided on the POC_Conf._Call_9-27-11 to use the PO_DBXref.txt file on the SVNto store the updated dbxrefs stanzas.

• An older version of this file is also available through the documentation page on the PO website:

LC: Do we really need to have this file PO_DBXref.txt on our website, since it is just for our internal use? If so, it should be updated. It could be a link to the SVN.

dbxref.txt file

The file on SVN only has correct text for the stanzas we have changed. Can JE give RW the text that AmiGO is actually using, so she can update dbxref.txt?

Link on http://plantontology.org/docs/docs.html goes old file. Should link to SVN?

 What parts of these stanzas does AmiGO actually use?  If JE could give us a copy of the info 
from the AmiGO file, we could compare the contents of the txt file and the actual functional links 

We decided the following:

JE will go through the dbxref.txt file and make sure that it is consistent with the contents of the AmiGO Perl file. He will remove any stanzas that we are not using currently. These can be added back later when we need them, from the GO version of the dbxrefs file or from the older version on the SVN.

We will change the link on the Documentation page to go to the current file on the SVN. This change will be implemented in the upcoming release.

Status of Dbxrefs that have been fixed:

• Link to SF trackers (OB_SF_PO):

Link is currently in the PO_DBXref.txt file on SVN, appears to be working on beta.

For example see thallus, PO:0030027

• MaizeGDB Links:

These look like they are working ok on the beta browser.

See comments below about the classical maize genes

• Gramene links:

The ones on the beta browser are working, not sure if the ones that were broken are there- need to load some test data on beta to make sure.

The ones changed in dbxref.txt are (as of 9-29-11, #152):

abbreviation: GR

database: Gramene: A Comparative Mapping Resource for Grains.

object: Protein

example_id: GR:P93436

generic_url: http://www.gramene.org/

url_syntax: http://www.gramene.org/db/protein/protein_search?acc=[example_id]

url_example: http://www.gramene.org/db/protein/protein_search?acc=P93436

Note: This URL actually refers to Gramene proteins. Should the stanza be renamed?

abbreviation: GR_MUT

database: Gramene: A Comparative Mapping Resource for Grains.

object: Mutants

example_id: GR_MUT:GR:0060198

generic_url: http://www.gramene.org/

url_syntax: http://www.gramene.org/db/genes/search_gene?acc=[example_id]

url_example: http://www.gramene.org/db/genes/search_gene?acc=GR:0060198

Note that this URL actually refers to Gramene genes and mutants now. Should the rest of the stanza be edited and renamed?

PJ will discuss the links with the Gramene group on the weekly conf call, to see if they would prefer to go to having all the database objects linked to the "Quick Search" page, rather than the individual links to each type of object. This would be consistent with the model GO is following of only linking to one universal url for each databse.

• Solanaceae Genome Network (SGN) links:

A couple of the links on live to SGN were also broken. We have been working wiht NM @SGN to get these straightened out.

The stanzas below were updated on the PO_DBXref.txt file on SVN, and appear to be working on beta.

However, many of the id’s from the association file, have no corresponding pages (e.g., id’s of 645 or below) and some of the id’s may not be going to the correct page.

For example, go to the Live PO page, search annotations for “LA0085” and then click on the link to go to the page for germplasm LA0085 in S. lycopersicum.

On this page, and in the association file, LA0085 has the id 4420, but if you go to http://solgenomics.net/stock/4420/view/, it is a dead link.

PO will update the link to use the new urls (see below:)

SGN should be sure that all germplasms have an ID for this url and that those id’s are correct. SGN should check and uptdate the stock/germplasm ids in all of the germplasm association file.

abbreviation: SGN_gene

database: Sol Genomics Network

object: locus identifier

example_id: SGN_gene:7740

generic url: http://solgenomics.net/

url_syntax: http://solgenomics.net/phenome/locus_display.pl?locus_id=[example_id]

url_example: http://solgenomics.net/phenome/locus_display.pl?locus_id=7740

abbreviation: SGN_germplasm

database: Sol Genomics Network

object: stock/accession/germplasm identifier

example_id: SGN_germplasm:6583

generic url: http://solgenomics.net/

url_syntax: http://solgenomics.net/stock/[example_id]/view/

url_example: http://solgenomics.net/stock/6583/view/

abbreviation: SGN_ref

database: Sol Genomics Network

object: reference identifier

example_id: SGN_ref:800

generic url: http://solgenomics.net/

url_syntax: http://www.sgn.cornell.edu/chado/publication.pl?pub_id=[example_id]

url_example: http://www.sgn.cornell.edu/chado/publication.pl?pub_id=800

Note: One of the common links that SGN uses (pub_id=861) goes to a page with basically no information. NM is working on fixing it.

For the next release, PJ suggested we should see if they would prefer to go to having all the database objects linked to something like a "Quick Search" page, rather than the individual links to each type of object. This would be consistent with the model GO is following.

• Links to the non-standard refs:

RW Created a PO_references page on the wiki and added a few of the PO_REF definition dbxrefs to the ontology file, for testing.

For example see: thallus (PO:0030027) working on beta browser

Annotation files

MaizeGDB

MS is working on updating the MaizeGDB annotation files that are on our SVN. They should updated by the end of the week.

Mappings to the Classical Mapped Maize Genes:

PJ raised a point last week that we should have the set of Maize classical genes mapped to the maize gene models.

Since MS is really busy right now, JP is going to insert the mappings in the association files based on Schnable's mapping data files. Once MS updates the association files, we can just rerun them, or let her do it at her end for later releases. This raised the issue of how to link to the dbxrefs and not end up with broken links.

From JP by email: I did some testing on PO beta, swapping in classical gene names & symbols for gene model id's (GRMZ*) on maize annotations. We wanted to see what the effect(s) would be if we replaced DB Object ID (col 2) with a locus id and/or replaced DB Object Symbol (col 3) with a classical maize gene symbol provided by CoGe.

JP created two new association rows in the test load file Justin Elser has been using in Beta to test MaizeGDB uploads to test these:

1. MaizeGDB 12768 wx1 - PO:0006340 PMID:20110808|MaizeGDB:9021423 IEP - A waxy1 GRMZM2G099328 gene_model taxon:4577 20110914 MaizeGDB part_of(PO:0009025)

2. MaizeGDB GRMZM2G099328 wx1 - PO:0006340 PMID:20110808|MaizeGDB:9021423 IEP - A waxy1 - gene_model taxon:4577 20110914 MaizeGDB part_of(PO:0009025)

(Don't worry about the fact that my chosen gene model id doesn't actually map to "wx1". That's irrelevant for our test.)

Next, JE reloaded his test annotation data on beta (thanks, Justin). Here are the results on the AmiGO browser:

(replaced col 2 with the locus id for "waxy1", moved the gene model id to synonyms)

(left the gene model id in col 2)

So the question on the table, I think, is which of these two approaches should we take? Please peruse these two pages and their links to MaizeGDB and let me know what you think. My gut feeling is that the first option (use locus id) is best for the users, but it will require add'l mapping work -- for us and for Mary -- to create an associated list of gene model id's to locus id's.

Before the conf call we discussed this and decided that option 2 was preferable.

Gramene:

From last meeting: New annotations should be coming from Gramene in the next 1-2 weeks.

PJ and LC have been in communication with KYC at Gramene and pointed them to the editor's file as the most current one. Not sure what the time line on these will be.

We are also working on adding new annotation files for Physcomitrella, grape and strawberry. We will need to add DBXrefs for these sources as well.

Physcomitrella

PJ has been in contact with SR and his collaborators. We are waiting for a few changes to be done and updates to the files.

LC will follow up with them and help them get the annotations sent in. Will need to set up and SVN login for someone there

strawberry

We have a set of annotations from strawberry- from last week's meeting: will link to the Roseacea database- need link, pj will send it to us.

Grape

- The original contributor, Nicola sent an additional file with the evidence code IEA (Inferred from Electronic Annotation) - thus these are based on homology to Arabidopsis, not actual assays in the grape leaf. Not a strong evidence to create annotations based on. PJ requested that they substitute the code: 'ISS'

The current file LC is are working on should be combined with the new one, put onto the SVN, but not loaded into the database. Gene or coding region may match, but without promoter analysis, it is too weak.

At last week's meeting, [POC_Conf._Call_9-27-11] we decided that we will not try to include these in this release, but will continue building it. Column 1. DB reference should be to CRIBI (sp?) link?. PJ will talk to PlexDB to get the microarray expression data- PlexDB has ~17 grape experiments, so we could link to m/a probe ids mapped to the genes. Julie Dickerson may be able to send us a dump for the database.

 Today, the decision was to include this information  Did I catch this correctly?

Final edits to be fixed:

• lemma and palea

PJ has requested that these be returned to being part_of spikelet floret, instead of spikelet. This is how most people think of them.

In order to do this (an maintain the logical consistency of the PO, lemma and palea would have to be is_a phyllome, instead of is_a flower bract.

We came up with the following solution:

Flower bracts are part of the flower, but not part of the perianth. Change def. of flower bract to say "subtends the perianth of a flower" instead of "subtends a flower" and make it part_of flower.

In this case, lemma and palea can be is_a flower bract and still be part_of spikelet floret. RW will update relations and modify definitions and comments accordingly. See SourceForge tracker for new definitions.

Make flower bract part_of flower.

Inclusion of Translations- Spanish and Japanese Synonyms of term names

The Spanish synonyms are now working fine, have the descriptor "(Spanish)" following them.

• Japanese synonyms?

Will have Japanese synonyms in obo file, but JE will strip them out before the file gets loaded on AmiGO

Japanese synonyms are not currently in the editors' file. Should be added. What is the status on this?

For the release, we will highlight this new feature as we are the first ontology to offer spanish (and possibly Japanese) translations. Will send out the announcements in Spanish as well.

JE- the code is written and needs to be tested. RW will send in the list of new terms to Yukiko to get them translated. New Spanish synonyms have been added to the Obo file.

• translation files should be updated on the SVN as they are changed.

Suggestion to automatically create tab-delimited text files with ID, name, definition, and translation and keep this on SVN. Translators could update this file directly, then editors could run JE's script to insert new synonyms with each release. Would need separate files for PAO and PGDSO (at least for now).

Plan for the next few weeks:

LC will be away at a course from 10-16 to 11-2. We want to continue having meetings- RW will host

Upcoming meetings and Presentations 2011/2012:

Plant Genomes & Biotechnology: From Genes to Networks, CSHL

Dates: November 30 - December 3, 2011 Abstract Deadline: September 9, 2011...

Academic/Student No-Housing Package $720

RW registered and submitted abstract (Media:PO_abstract_PGBatCSHL2011.pdf). Will need to present a preliminary data analysis for the talk/poster.

PAG 2012

January 14-18, 2012, San Diego, California

• Registration and Abstract submissions: open on Sept 22
• Early Pre-Registration Deadline: October 31
• Travel Grants Deadline: November 1
• Abstract Submission Deadline: November 11

Our proposal for an Ontology workshop for PAG has been approved. It will be on Saturday January 14th from 10:20am-12:30pm

Use of Ontologies for Organizing Plant and Animal Genomics Data

Suggestions for speakers??

Should have talks from people working on ontology applications, rather than ontology development, to show utility.

LC is presenting in the Plant Phenotypes workshop on Sunday Morning, 15 January 2012 -- 8:00 am - 10:10 am.

The PO will take part in an Outreach booth organized by MaizeGDB

Other activities?

Phenotype RCN meeting, 23-25 February 2012

The dates: February 23-25, 2012 (Thursday, Friday, 1/2 Saturday) have been confirmed for the next annual Phenotype RCN meeting.

It will be held again at NESCent (Durham, NC).

RW has a friend there she can stay with and is interested in going.

5th International Biocuration Conference

April 2-4, 2012, Washington DC

Call for abstracts is now open: Need to clarify the deadlines

There are three submission categories for abstracts:

1. Talk or Poster (with consideration for oral presentation)

2. Poster only

3. Workshop only

• Submission deadline November 30, 2011

• Notification of acceptance February 3, 2012

Instructions

There are seven topic sessions from which submitters are invited to select:

  1. Ontologies, standards and best practices, including gold standard datasets.
  2. Protein annotation; sequences, structures and pathways.
  3. Community annotation and Wikis.
  4. Genomics and metagenomics data curation.
  5. High throughput proteomics data (focus on NGS and MS data) curation and presentation.
  6. Literature collection, text mining and curation.
  7. Tools to assist curation, including automated pipelines. 

There are four submission tracks:

  1. Paper, with consideration for oral presentation
  2. Talk
  3. Workshop
  4. Poster 

PJ planning to attend and will be running a biocuration workshop, LC and RW can go, DWS will be away,

'From 9-27-11: PJ: we should ask MS, and possibly SR, to contribute to the abstract and the annotation guide.

We ran out of time to discuss this- tabled for next week

Botany 2012

July 7 - 11, 2012 - Columbus, Ohio

Call for Symposia, Colloquia and Workshops:

Submission site now open. Deadline for submissions is Saturday, October 15th, 2011 at 11:59 PM PST.

RW, DWS and MAG will put together a proposal for a half day hands-on workshop. The goal will be to teach people (mostly botanists) how to access and use the PO, including how to send feedback, suggest new terms, etc.

PJ: suggest that we go there with a 'draft' version of the Plant Phenotype Ontology and show them how to use these in character matrixes. RW will put the draft plant PATO on the SVN.

• We need to have a section sponsor for the workshop proposal, DWS suggested the structural section (Head: Bruce Kirchoff), paleo section or systematists. Development and structural, genetics, and teaching sections would also be appropriate co-sponsors.

From meeting website:

When you submit your proposal, you will be asked for the following information:

• What Societies and Sections will be sponsoring the session. Please provide contact information for any persons who have approved the sponsorship
• The societies participating in Botany 2012 of which you are a member
• The names, email and street addresses of the organizers, and which of them will lead the session
• A title and description of the session
• Keywords and web links that are relevant
• Requirements, such as room type, food, special equipment or transportation, depending on the type of proposal
• An estimate of anticipated attendance
• A list of invited speakers or leaders, if relevant.

We should also consider hosting an outreach booth.

Not a bad deal for non-profits: $500 for A 10 x 10 Booth Space at Botany 2012, and 2 complimentary registrations for the conference. (plus all the extras!)

• 2 months of Rotating Banner Ads in the online American Journal of Botany

• A Rotating Banner Ad in one edition of the online Plant Science Bulletin

• A Rotating Banner Ad on the Botany 2012 abstract submission site

• A Rotating Banner Ad on the 2012 Conference Registration site.

PJ will check with Gramene and xxxx to see if they want to co-host a booth.