MGED Board
 Advisory Board
 Board Meetings
 First Meeting
 Second Board
 Third Board
 Fourth Board
 Fifth Board
 Sixth Board
 Fifth Board Meeting
 Home  Meetings  Workgroups  Mission  MGED Board  Site Map

Home : MGED Board : Board Meetings : Fifth Board Meeting

Tues 11/26/02 conference call MGED board meeting

Board members attending:
1) Alvis Brazma
<brazma@ebi.ac.uk>
    (president)
2) Helen Causton
<helen.causton@csc.mrc.ac.uk>
3) Gavin Sherlock
<sherlock@genome.stanford.edu>
4) Paul Spellman
<spellman@bdgp.lbl.gov>         
    (treasurer)
5) Ronald Taylor
<Ronald.Taylor@UCHSC.edu>
    (secretary)
6) Cathy Ball
<ball@genome.stanford.edu>
7) Helen Parkinson
<parkinson@ebi.ac.uk>
8) Charles Troup
<charles_troup@agilent.com>
9) Chris Stoeckert
<stoeckrt@SNOWBALL.pcbi.upenn.edu>
10) Barbara Jasny
<bjasny@aaas.org>
 

Board members absent:
1) Terry Gaasterland
<gaasterl@cleo.rockefeller.edu>
                     - also:
tgbb@rockefeller.edu
2) Jason Goncalves
<jason.goncalves@iobion.com>
3) John Quackenbush
<johnq@tigr.org>
4) Martin Ringwald
<ringwald@informatics.jax.org>
5) Jason Stewart
<jason@openinformatics.com>
     - now departed from board
6) Yoshio Tateno
<ytateno@genes.nig.ac.jp>
    (MGED 5 host & organizer)
7) Pascal Hingamp
<hingamp@ciml.univ-mrs.fr>
 

Nominated board members for term starting in May 2003 - attending:
1) Michael Miller, Rossetta Biosoftware
michael_miller@rosettabio.com

Nominated board members for term starting in May 2003 - absent:
1) Kate Rice
cmr@sanger.ac.uk
2) Kousaku Ohkubo
kousaku@bioreg.Kyushu-u.ac.jp
3) Steve Chervitz
Steve_Chervitz@affymetrix.com
 

-----------------------------------------------------

11:05 AM EST

Paul formally opened the meeting. We proceeded through the agenda items sent out by Alvis.

--------

Item: Legal issues and Paul's meeting with lawyers
1a. (possibly) reorganizing the board of directors - Alvis, Paul

Paul: I meet and signed a contract with our attorneys, Silk, Adler, & Colvin. They have got us a number for a bank account. I have asked Terry to send them a retainer check. The firm will send us bills every month or so. This has gone well.

The time frame for getting all of our documentation, including the letter from IRS saying we are a tax-exempt organization, etc is about six more months. (It takes the IRS a long time to do stuff.)

Ron: what about the subject of using email as official notification of board meetings?

Paul: yes, they said it is possible. I will press it.

Paul: Silk, Adler have pointed out to me that we wrote something into our bylaws regarding quorum size which is illegal. Our minimum quorum size is too small. We will have to increase it.

Paul: that's about it on this subject. I will have the rest of the documents they need this week. Then Silk, Adler will plug away. Hopefully sometime in Jan-Feb we will start to get responses from the IRS.

Charles: what will be the new quorum size?

Paul: By law, a minimum quorum size is 1/5th of the total number of board members. Our lawyers suggest a higher minimum. With our current board size of 16, a min of 1/5th would mean 4 board members.

Paul: our lawyers think 1/5th is a bit small. They were concerned about it. They would like to see the min quorum size moved up to 1/2 of the total board size.

Alvis: so the alternatives for a quorum are
a) keeping 1/5th as the minimum quorum, a quorum would be 5 out of an expanded 25 member board
b) moving the min quorum size up to 1/2, as recommended our lawyers. If we do that, then we might change the way in which the board works. We could have 10 official directors, which are rotated yearly out of the expanded board of 25. Using 1/2 as the min, we would then still need only 5 people to make up a quorum.

Paul, Alvis: we would need more of a committment to attend meetings from the 10 people selected for that year.

Alvis: I don't know which option is better. Both are viable alternatives.

Michael: what is a typical size of a board for a nonprofit organization?

Paul: our lawyers thought 25 was rather large - mainly because they recommend at least half the board show up. They would prefer something like 9 to 13 board members. Basically, they want to ensure that a larger fraction participates in decision-making at the board meetings.

Alvis: do we have to make a decision now?

Paul: no, we don't have to make a board size decision now.

Alvis: I agree, this can wait until the board elections in May, and we can talk more in Denver. By May we should have all 25 board members selected, and we can talk about how we want to operate formally. Informally, I don't believe this should make much difference. The big difference, if we go with 10 official directors, is that they would have to make a committment to participate in at least every other conference call.

Paul: I think the selected ten should make a committment to make nearly every conference call.

Alvis: OK, we will talk in Denver and make a decision by May.

Board: agreed.

--------

Item: ISMB 03 tutorial proposals

Alvis: Neither John Q nor Jason Stewart is present, so I will start of the discussion. There are two possible tutorials - one for MAGE-ML, and one for normalization. Jason S has previously volunteered to put together a proposal for MAGE-ML. But time is running short, so maybe I had better urge on Ugis to put a proposal together, since Ugis will go to ISMB-03. The submission deadline is next week, not long from now.

Chris: On the ISMB web site, the tutorial submission deadline is given as Dec 3.

Michael: I cannot go, but I can help Ugis with the proposal.

Alvis: good. I will ask Ugis if he can do this. The proposal length is only about one page. Ugis can give a 2 hr tutorial if he does it alone, a 4 hr tutorial if he does it with a partner. I will find money for Ugis's travel to ISMB-03.

Alvis: I don't know the status of the normalization tutorial. Cathy or Gavin, could you fill us in?

Cathy: I have not heard from John Q since I forwarded him the ISMB tutorial proposal info. I will call him to find out what is going on.

Gavin: I have not heard anything additional.

Alvis: Ok, please check with John. I understand that ISMB does not want to pay for travel for the tutorial speakers. Whether MGED has enough money by that time to contribute something is not clear, so travel decisions cannot be based on that. I will find money for Ugis from my grants.

Cathy, Gavin: John Q should be able to go; it may just be John. We cannot commit to the travel right now.

Alvis: Gavin or Cathy, can you contact John Q re putting the normalization tutorial proposal together?

Gavin, Cathy: OK.

--------

Item: NIST meeting

Alvis: I forwarded an email on this to everybody. John Q will be there. Does anybody else want to go? I don't think there is any big need, but you are welcome to attend. ($60 for registration.)

Gavin: if John is going, that is fine.

Alvis: OK, we will delegate this to John.

--------

Item: MGED 6 issues

Alvis: Pascal is not here. He sent one question for us in regard to MGED 6 planning:

Do we pay something additional in order to get the conference company to handle registration payments via credit card?

Paul: yes, I use a credit card for conference registration myself, for example.

Everybody: yes.

Alvis: OK, I'll tell him the board says "yes". The difference in cost is small.

--------

Item: Ontology meeting decisions at EBI

Alvis: Chris, Paul, Helen, Martin Ringwald, and myself attended, among others. Chris, can you give a report on this?

Chris: Before the meeting, Alvis had asked me to write a position statement. I did so. The statement basically states what is the point of the MGED ontology we have developed. It is designed to help annotators annotate microarray experiments, and to help software developers write tools to allow people to perform such annotation.

<Chris then described the position statement in more detail. The full statement is appended at the bottom of the minutes.>

Chris: One of the things spoken of in the position statement is the development of a core, stable ontology that will not change. To bring in new terms, and adjust to new technologies, there will be an extended ontology will be layered on top of the core ontology. The PDB uses such a model for their ontology describing protein structures, and it appears to work well. In discussing this with folks, there seemed to be real support for this approach at the meeting, especially in how the ontology would integrate with the MAGE effort. (The core ontology would be designed to fit seamlessly with MAGE 1.0. That is, the core would help users fill in fields required by MAGE and MIAME.)

The extended ontology will continue to grow.

Chris: So the conclusion from the meeting is that the core ontology will work with the existing MAGE standard. We will change the terms and concepts so that the ontology will fit MAGE 1.0, and we will continue working with the MAGE folks.

Michael: will there be revisions of the core ontology as time moves on? Very clear cut, like a 2.0 release?

Chris: the notion is that the core will work with the existing MAGE 1.0 standard. When MAGE 2.0 comes out, there will certainly need to be a new release of the core ontology, to remain in sync with MAGE.

Chris: there is a target date for production of this core ontology. We want to get this done by September, by MGED 6 at the latest. Helen P and I will have ontology jamborees. Paul has volunteered to participate. In next couple months, I hope to have a draft of the core ontology which I will post to the MAGE and ontology mailing lists. Then we will make revisions.

Alvis: we have two parts/stages:
a) development of a core ontology to fit MAGE 1.0, to be done by Sept.
b) MAGE 2 development and ontology development will have to coordinate very closely

Chris, Cathy: yes, agreed.

Helen P: use of proprietary ontologies was discussed at the meeting. I believe the feeling at the meeting was that we would not recommend the use of such ontologies to annotate data. Is this something the board would like to discuss?

Kathy: the other thing discussed was having the ontologies group become part of GOBO.

Helen P: we are, in effect, already part of GOBO. Our ontology is on the GOBO site. But we might want to declare something formally.

Chris: I believe we did put that in our mission statement.

Ron: so ... going back to proprietary, do we want to formally discourage the use of proprietary ontologies?

Paul: what exactly would that mean?

Chris: we can say that, in the specification of external resources in the MGED ontology, we will not specify any ontologies that are not open.

Paul: yes, that is a good, clear statement.

Charles: would ArrayExpress accept proprietary terms?

Paul: yes. The only point is that those terms would not be in the MGED ontology.

Alvis: if there is a submission to ArrayExpress using a proprietary (or completely private) ontology, it will be accepted. We can just mark it as using a non-MGED-approved ontology, implying that analysis of that data set will be questionable.

Helen P: Ok, at the Nov. 17-20 Standards and Ontologies for Functional Genomics (SOFG) Conference itself there was discussion on resolving some of the differences in the parallel ontologies coming out of the anatomy and gene expression worlds. I think Win Hyde (?) was the driving force behind that. If anybody is interested, they can join in that effort.

Alvis: just to re-emphasize what Helen said - it was a positive outcome. Win Hyde, Martin Ringwald, Duncan Davidson and several others participated.So this effort may take off.

Chris: another exciting thing that may happen is the creation of a central site for open ontologies.

Michael: there are bound to be overlapping ontologies. The same concept may have two different names, for example. An central repository may allow some group to create and store mappings between them, which would be nice.

Charles: that is one of the beauties of having a central site. We can have a set of aliases that people will be happy with. And hopefully most of the community will be satisfied by that database.

Chris: should I send out a modified ontology position statement, in light of the discussions that happened at the meeting and today?

Helen: yes, send it to Ron, to put in the minutes.

Ron: OK, I'll expect that.

Alvis: and put it up on the web site.

--------

Item: Other high throughput technology meeting at EBI

Paul: It was mainly a meeting with the HUPO human proteome guys. I was very disappointed. They told us what they were doing without asking for any input from us. Nothing there, unless they get far more organized.

Charles: who were these people?

Paul: the Human Proteome Organization. The lead guy was from EBI.

Helen C: they did not understand the extant of what they were trying to do.

Paul: right, they were not arrogant, just lost. They did not have a well-thought-out agenda.

Helen C: they are actually having a meeting at Versailles this week. Maybe they will do something useful there. There are some people from Glasgow who have modified MAGE-ML to handle mass spec data who will be there.

Chris: from the mass-spec standpoint, the HUPO people do not seem to be involving the major players. That is of concern.

Alvis: so the conclusion is we wait on this - we can't do anything on this until they mature. So, we agree that the decision on this is not to do anything right now?

Board: agreed.

Chris: they will have another meeting in January at the EBI.

Alvis, Helen C: we can have somebody pop in to keep us apprised.

%%%%%

Charles: I have something I would like to add. I have started to draft an RFP on query service for MAGE 1.0. The RFP for that will hopefully get approved in January by the OMG/LSR. Actual proposals would then be due in 6-7 months after that.

Alvis: please send a draft to the board.

Gavin: this RFP can be discussed at the Dec programming jamboree.

Charles: I will try find the time to do it before then - I will try to get the RFP out this week.

Gavin: one more thing re MGED 6 planning- I would like to include the BioConductor people. Perhaps include a BioConductor tutorial - it is an open source R project gaining in popularity in array analysis. Perhaps a tutorial on R, too.

Helen C: that would be great.

Alvis: I propose for next phoning that we try to have a very rough draft of the MGED 6 program, for discussion purposes. We can include something on the BioConductor group in that draft.

Gavin: fine.

Alvis: a couple more minor things.
1) Our solicitation letter will soon go to EBI industry partners - about 20 companies.
2) I have not found a good person at Lion to whom to give our letter. If anybody has a suggestion, let me know.
3) I suggest that our December phone conference should be a week later than usual, that is, in the first week in January, to get past the holidays.

Board: agreed to have the next conf call board meeting on Tuesday Jan 7th, 2003 at 11 AM EST.

Ron: two conference calls in January, then?

Alvis: yes.

Paul: I will send the meeting sound file to Ron.

Alvis: OK, happy new year and I will see everybody (online) in January.

Meeting closed at 12 noon.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%

from Chris Stoeckert:

The position statement and first draft of the core ontology have been posted at the ontologies site (http://mged.sourceforge.net/ontologies/index.php).

MGED Ontology Position Statement

        The primary purpose of the MGED Ontology is to provide standard terms for the annotation of microarray experiments. These terms will enable structured queries of elements of the experiments. Furthermore, the terms will also enable unambiguous descriptions of how the experiment was performed. The terms will be provided in the form of an ontology which means that the terms will be organized into classes with properties and will be defined. A standard ontology format will be used. For descriptions of biological material (biomaterial) and certain treatments used in the experiment, terms may come from external resources that are specified in the Ontology. Software programs utilizing the Ontology are expected to generate forms for annotation, populate databases directly, or generate files in the established MAGE-ML format.  Thus, the Ontology will be used directly by investigators annotating their microarray experiments as well as by software and database developers and therefore will be developed with these very practical applications in mind.

As the terms encoded in the Ontology ultimately will be placed in a MAGE-ML document, to facilitate this process it is important that the efforts of the MAGE and Ontology working groups are coordinated where there is overlap. Ontology classes will strive to maintain the same names and relationships where appropriate to MAGE classes. For instance, 'Experiment' is used because of MAGE in place of 'Study' even though 'Experiment' is not ideal as it interpreted in multiple ways including the distinction from observational studies. As the MAGE effort moves forward and requires terms to describe elements of microarray experiment, the Ontology group should be consulted so as not to duplicate efforts and to use the Ontology to establish terms of choice.

The Ontology is expected to continually grow especially as new applications of microarray technology arise that require descriptive terms.  For software programming efforts a fixed Ontology is desired so that usage need not be constantly revised. In order to meet both needs, a core MGED Ontology will be established that will not change. A second layer of the Ontology, the extended MGED Ontology, will contain all additional terms that must be logically consistent with the core Ontology. This approach follows the successful model of the Protein Data Bank and the core mmCIF dictionary approved by the IUCr and exchange mmCIF dictionary.

        The core MGED ontology will made to work seamlessly with the published MAGE 1.0 schema and will be completed no later than Sept. 3, 2003 so that it can be presented and voted on at the MGED 6 meeting. To achieve this goal, a series of MGED Ontology Jamborees will be held in the same spirit as the MAGE programming jamborees. The extended MGED ontology will be developed in conjunction with MAGE efforts to develop MAGE 2.0.

        The MGED Ontology is one of several ongoing efforts to represent the domains of genomics and proteomics. A global open biological ontologies effort (GOBO) is underway to collect freely-available ontologies that use a standard format and are orthogonal in their coverage. The MGED Ontology will participate in this effort through the provision of terms for experimental
conditions. Furthermore, to advance the use of freely available ontologies as part of the GOBO mission, the MGED Ontology will only include reference to freely available resources.

 

Home | Meetings | Workgroups | Mission | MGED Board | Site Map
 

Last modified: 26 Sep, 2005.                              Contact Us
This site is hosted by the EMBL -
European Bioinformatics Institute
The maintenance of these pages is partially supported by the European Commission as part of the
TEMBLOR project.
 

MGED Sponsors: