Managers 28Mar07: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
mNo edit summary
mNo edit summary
Line 1: Line 1:
GO Managers, Weds. February 28
GO Managers, Weds. February 28<br>
8 AM PDT / 10 AM CDT / 11 AM EDT / 4 PM BST
8 AM PDT / 10 AM CDT / 11 AM EDT / 4 PM BST<br>




Line 19: Line 19:
Action Item: Chris will try to automatically mine biochemical pathways from Reactome, IMG, etc .
Action Item: Chris will try to automatically mine biochemical pathways from Reactome, IMG, etc .


David reports for Chris: Chris tried with reactome and has a tool for this.
David reports for Chris: Chris tried with reactome and has a tool for this.<br>
He had some problems with conversion of reactome terms to go terms. Perhaps he could do a webex demo of the tool?
He had some problems with conversion of reactome terms to go terms. <br>Perhaps he could do a webex demo of the tool?


Action RC: group of evelyn, jb, cm, jen clark to explore possible cancer grant proposal
Action RC: group of evelyn, jb, cm, jen clark to explore possible cancer grant proposal
Line 35: Line 35:
(Jennifer Clark, Val Wood, Michelle Gwinn and Ian Paulsen are overhauling transporter activity terms on conference calls that occur once a week.)
(Jennifer Clark, Val Wood, Michelle Gwinn and Ian Paulsen are overhauling transporter activity terms on conference calls that occur once a week.)


Transport work is progressing well with the (draft) graph improving very rapidly.  
Transport work is progressing well with the (draft) graph improving very rapidly. <br>
Jen: How does the annotator community feel about the large number of obsoletions proposed.  
Jen: How does the annotator community feel about the large number of obsoletions proposed. <br>
All: No problems have beenreported. It seems as the though the terms being obsoleted are all bad terms that were not used so will not be missed.  
All: No problems have beenreported. It seems as the though the terms being obsoleted are all bad terms that were not used so will not be missed.  


Line 50: Line 50:
'''iv. Ontology development newsletter/mailing list.'''
'''iv. Ontology development newsletter/mailing list.'''


Jen received a question from a person who signed up to the ontology development working group and is wondering why they have not been contacted.  
Jen received a question from a person who signed up to the ontology development working group and is wondering why they have not been contacted.<br>
Q: will there be an ontology development mailing list or regular conference call?
Q: will there be an ontology development mailing list or regular conference call?


Line 61: Line 61:


Midori: It's hard to think of a clear group who would be on this mailing list. There are many people who might be interested in content development and they're all on the GO list so  
Midori: It's hard to think of a clear group who would be on this mailing list. There are many people who might be interested in content development and they're all on the GO list so  
it makes sense to send information there.  
it makes sense to send information there. <br>
Also if we do have a smaller group of people who are interested in ontology development and we put them on an ontology development mailing list then there will be many mails going
Also if we do have a smaller group of people who are interested in ontology development and we put them on an ontology development mailing list then there will be many mails going
to the list that are not relevant to their area of interest. This would be a waste of their time.  
to the list that are not relevant to their area of interest. This would be a waste of their time.  
Line 74: Line 74:


1) What are your areas of biological expertise? We will be calling on everyone to join the small conference calls/content meetings and it would be very useful to have  
1) What are your areas of biological expertise? We will be calling on everyone to join the small conference calls/content meetings and it would be very useful to have  
a clear idea of which topics each individual can contribute to.
a clear idea of which topics each individual can contribute to.<br>
2) When do you have time to contribute? Each person has quiet periods and more busy times when they may be travelling or writing grants. It would be very useful to know
2) When do you have time to contribute? Each person has quiet periods and more busy times when they may be travelling or writing grants. It would be very useful to know
when people are less busy so we can call on them to contribute.  
when people are less busy so we can call on them to contribute.<br>




'''v. MIT Meeting with Midori Harris, David Hill and Jane Lomax.'''
'''v. MIT Meeting with Midori Harris, David Hill and Jane Lomax.'''


A group at MIT have been using the annotations to analyse the GO graph. They look at which terms have far more or far fewer annotations and use this to work out where we should be concentrating our efforts to expand the graph. Midori, David and Jane have been to visit and talk with them.  
A group at MIT have been using the annotations to analyze the GO graph. They look at which terms have far more or far fewer annotations and use this to work out where we should be concentrating our efforts to expand the graph. Midori, David and Jane have been to visit and talk with them.  


Meeting was good! We are now making some changes to the graph to improve it based on their suggestions.  
Meeting was good! We are now making some changes to the graph to improve it based on their suggestions.  
Line 102: Line 102:
'''i.    New EBI training room.'''
'''i.    New EBI training room.'''


The EBI in Hinxton, UK is having an extension built and this will include a training room with 40 desktop computers and space for two people at each. The GO Editorial Office and GOA groups will be able to use this free of charge for external training. This is raised as a possible venue for future GO training events. It is scheduled to be ready by June but current information suggests that we should not count on it until september.  
The EBI in Hinxton, UK is having an extension built and this will include a training room with 40 desktop computers and space for two people at each. The GO Editorial Office and GOA groups will be able to use this free of charge for external training. This is raised as a possible venue for future GO training events. It is scheduled to be ready by June but current information suggests that we should not count on it until September.  


Rex: We had considered the possibility of a reference genome annotation meeting at the EBI. This room would be good for that.   
Rex: We had considered the possibility of a reference genome annotation meeting at the EBI. This room would be good for that.   
Line 110: Line 110:


ii.  Ranjana Kishore will be attending the International worm meeting at    
ii.  Ranjana Kishore will be attending the International worm meeting at    
UCLA, Los Angeles and giving GO talk.  
UCLA, Los Angeles and giving A GO talk.  


'''iii. BioSapiens '''
'''iii. BioSapiens '''
Line 116: Line 116:
Question from Evelyn Camon via Jennifer:
Question from Evelyn Camon via Jennifer:


GOA attended a curator meeting today where Michele Magrane of UniProtKB(EBI) gave a talk about the Biosapiens project(presentation on line internally at EBI).
GOA attended a curator meeting today where Michele Magrane of UniProtKB(EBI) gave a talk about the BioSapiens project(presentation on line internally at EBI).


This project involves 64 DAS servers and 19 institutes. The project is focussed on the large scale annotation of genome data. They are trying to resolve the problem of the inconsistency of feature types between DAS servers. There is too much data with too little structure and some data appears to replicate the same annotation. For example, data originating from SMART feeds into InterPro, and then UniProt, giving the user the idea that there are 3 pieces of supporting evidence when in fact there was only one. So the Consortium is creating a protein sequence feature ontology. It was decided not to use the Sequence Ontology as that was focusing on gene features, but it was suggested that maybe someday they could be linked. When Evelyn asked if this new ontology (which btw includes Subcellular location) will be available in OBO, the anwser was not until it is complete OR until the Consortium agree to release it.
This project involves 64 DAS servers and 19 institutes. The project is focussed on the large scale annotation of genome data. They are trying to resolve the problem of the inconsistency of feature types between DAS servers. There is too much data with too little structure and some data appears to replicate the same annotation. For example, data originating from SMART feeds into InterPro, and then UniProt, giving the user the idea that there are 3 pieces of supporting evidence when in fact there was only one. So the Consortium is creating a protein sequence feature ontology. It was decided not to use the Sequence Ontology as that was focusing on gene features, but it was suggested that maybe someday they could be linked. When Evelyn asked if this new ontology (which btw includes Subcellular location) will be available in OBO, the anwser was not until it is complete OR until the Consortium agree to release it.


Q: Has any member of GOC been consulted on the Biosapien project?  
Q: Has any member of GOC been consulted on the Biosapiens project?  


===c. Software===
===c. Software===
Line 126: Line 126:
OBO-Edit 1.1 is still in testing phase. John Day-Richter  is now 2/3 time on version 2.0 because he gets 2/3 funding from NCBio.
OBO-Edit 1.1 is still in testing phase. John Day-Richter  is now 2/3 time on version 2.0 because he gets 2/3 funding from NCBio.


Mike: What is happening about adding tags for replacement terms into OBO format. Jane: We are still not there yet. We are awaiting the release of OBO-Edit 1.1.  
Mike: What is happening about adding tags for replacement terms into OBO format. Jane: We are still not there yet. We are awaiting the release of OBO-Edit 1.1. <br>
We'll contact mike when that happens.  
We'll contact mike when that happens.  


===d. Reference Genomes===
===d. Reference Genomes===


Not much to say this week.
Not much to say this week.<br>
There is about to be an e-mail about the new April target genes (Monday).  
There is about to be an e-mail about the new April target genes (Monday).  


Line 139: Line 139:
We are trying to merge the front pages of AmiGO and the GO website but it is proving difficult as they evolve independently. We are looking at mockups.  
We are trying to merge the front pages of AmiGO and the GO website but it is proving difficult as they evolve independently. We are looking at mockups.  


Possible go-help improvements:
Possible go-help improvements:<br>
Currently Mike is the moderator and eliminates a lot of spam from the list. In addition there is one person on duty each fortnight who makes sure that all mails are answered and forwards them to groups that might be best placed to answer them. Jane and Eurie ask if it would be good if all mails to go help went to a single gmail account that we could all access that the person on duty also did the spam filtering. Mike says this would not be a great idea as the spam filtering is quite subtle sometimes and he is very well practised at it and may as well go on doing it. Also there is no reason to burden the duty person with this as he is happy doing it.  
Currently Mike is the moderator and eliminates a lot of spam from the list. In addition there is one person on duty each fortnight who makes sure that all mails are answered and forwards them to groups that might be best placed to answer them. Jane and Eurie ask if it would be good if all mails to go help went to a single gmail account that we could all access that the person on duty also did the spam filtering. Mike says this would not be a great idea as the spam filtering is quite subtle sometimes and he is very well practised at it and may as well go on doing it. Also there is no reason to burden the duty person with this as he is happy doing it.  


Line 159: Line 159:
Jen: If we do go then instead of getting people to do academic talks couldn't we do a training day to teach microarray users how to use GO? We could do this jointly with one of the major tool developers. If we are paying for someone to go then Fiona McCarthy might be a good candidate as she is from Australia and goes there regularly anyway.  
Jen: If we do go then instead of getting people to do academic talks couldn't we do a training day to teach microarray users how to use GO? We could do this jointly with one of the major tool developers. If we are paying for someone to go then Fiona McCarthy might be a good candidate as she is from Australia and goes there regularly anyway.  


Action: No decision was made on whether to send anyone.  
Action: No decision was made on whether to send anyone. <br>
Jen will ask Fiona if she would eb interested in principle.  
Jen will ask Fiona if she would eb interested in principle.  




'''Institute of Animal Health (IAH): oxford. '''
'''Institute of Animal Health (IAH): oxford. '''
They have asked for a GO course.  
They have asked for a GO course. <br>
Evelyn Camon did her PhD there so is interested. She has suggested a double act with Jennifer.  
Evelyn Camon did her PhD there so is interested. She has suggested a double act with Jennifer.  



Revision as of 08:34, 29 March 2007

GO Managers, Weds. February 28
8 AM PDT / 10 AM CDT / 11 AM EDT / 4 PM BST


Participants: Midori Harris, Jennifer Clark, Rex Chisholm, David Hill, Eurie Hong, Mike Cherry.

Agenda/Chair/Minutes: Jennifer Clark



1. Review Action Items from previous meeting

Action Item: Crediting contributors: how to reference individuals (Ontology Development, User Advocacy)

This is ongoing. It can be removed from the list of action items until progress is reported.

Action Item: Chris will try to automatically mine biochemical pathways from Reactome, IMG, etc .

David reports for Chris: Chris tried with reactome and has a tool for this.
He had some problems with conversion of reactome terms to go terms.
Perhaps he could do a webex demo of the tool?

Action RC: group of evelyn, jb, cm, jen clark to explore possible cancer grant proposal

No progress on this.


2. Individual Manager reports

a. Ontology Development

i. Transporter activity overhaul.

(Jennifer Clark, Val Wood, Michelle Gwinn and Ian Paulsen are overhauling transporter activity terms on conference calls that occur once a week.)

Transport work is progressing well with the (draft) graph improving very rapidly.
Jen: How does the annotator community feel about the large number of obsoletions proposed.
All: No problems have beenreported. It seems as the though the terms being obsoleted are all bad terms that were not used so will not be missed.

ii. Plant terms sensu spin-off

The conference call to fix plant sensu terms went very well and so we are continuing with once a week calls to try to resolve some very old plant sourceforge items.

iii. RNA processing

The RNA processing part of the graph is currently being worked on.

iv. Ontology development newsletter/mailing list.

Jen received a question from a person who signed up to the ontology development working group and is wondering why they have not been contacted.
Q: will there be an ontology development mailing list or regular conference call?

Discussion:

Midori: There is information for people on the working group and there always has been. There is sourceforge and the GO mailing list and now there is the Wiki. It's just that for some reason people are not managing to keep up with reading that information.

Jen: Should we have a mailing list just for the ontology working group?

Midori: It's hard to think of a clear group who would be on this mailing list. There are many people who might be interested in content development and they're all on the GO list so it makes sense to send information there.
Also if we do have a smaller group of people who are interested in ontology development and we put them on an ontology development mailing list then there will be many mails going to the list that are not relevant to their area of interest. This would be a waste of their time.

David: Our system of ontology development is changing so that instead of discussing all decisions on sourceforge we are gathering all the relevant experts together and having a small content meeting and just making the changes there and then. Some people have said that they don't like this as they say we are just making unilateral decisions without letting group comment. However, this is not the case, as we ensure that we have all the relevant people assembled, and we take all opinions into account. The new system is much quicker, and we are minuting all changes on the wiki.

For those people who volunteered to be on the ontology working group we would be very keen to know the following:

1) What are your areas of biological expertise? We will be calling on everyone to join the small conference calls/content meetings and it would be very useful to have a clear idea of which topics each individual can contribute to.
2) When do you have time to contribute? Each person has quiet periods and more busy times when they may be travelling or writing grants. It would be very useful to know when people are less busy so we can call on them to contribute.


v. MIT Meeting with Midori Harris, David Hill and Jane Lomax.

A group at MIT have been using the annotations to analyze the GO graph. They look at which terms have far more or far fewer annotations and use this to work out where we should be concentrating our efforts to expand the graph. Midori, David and Jane have been to visit and talk with them.

Meeting was good! We are now making some changes to the graph to improve it based on their suggestions.

There were some things they hadn't taken into account like the effect of the difference in the part_of and is_a relationships. We explained that to them, and they are going to incorporate is and rerun their analysis. We also explained some aspects of annotation and differing levels of completeness of annotation for different species and they will take this into account now.

They showed where there were some terms that needed some work and we are now fixing those.

Terms were too general or too specific (Ed: I do not completely get this, can someone expand? Do you mean terms were correctly defined but in the wrong place in the graph?)

We have started work on the too specific ones and the too general ones will be done next.

Once we have finished these improvements they will rerun the analysis. They will also rerun with different versions of the go over time. This may show metrics for improvements.

Summary: A good meeting with helpful, responsive collaborators and we will carry this on.


b. Outreach

i. New EBI training room.

The EBI in Hinxton, UK is having an extension built and this will include a training room with 40 desktop computers and space for two people at each. The GO Editorial Office and GOA groups will be able to use this free of charge for external training. This is raised as a possible venue for future GO training events. It is scheduled to be ready by June but current information suggests that we should not count on it until September.

Rex: We had considered the possibility of a reference genome annotation meeting at the EBI. This room would be good for that. The only difficulty may be finding a time when the accommodation at Hinxton is not full. Will this be a problem? Response: probably not.

WormBase

ii. Ranjana Kishore will be attending the International worm meeting at UCLA, Los Angeles and giving A GO talk.

iii. BioSapiens

Question from Evelyn Camon via Jennifer:

GOA attended a curator meeting today where Michele Magrane of UniProtKB(EBI) gave a talk about the BioSapiens project(presentation on line internally at EBI).

This project involves 64 DAS servers and 19 institutes. The project is focussed on the large scale annotation of genome data. They are trying to resolve the problem of the inconsistency of feature types between DAS servers. There is too much data with too little structure and some data appears to replicate the same annotation. For example, data originating from SMART feeds into InterPro, and then UniProt, giving the user the idea that there are 3 pieces of supporting evidence when in fact there was only one. So the Consortium is creating a protein sequence feature ontology. It was decided not to use the Sequence Ontology as that was focusing on gene features, but it was suggested that maybe someday they could be linked. When Evelyn asked if this new ontology (which btw includes Subcellular location) will be available in OBO, the anwser was not until it is complete OR until the Consortium agree to release it.

Q: Has any member of GOC been consulted on the Biosapiens project?

c. Software

OBO-Edit 1.1 is still in testing phase. John Day-Richter is now 2/3 time on version 2.0 because he gets 2/3 funding from NCBio.

Mike: What is happening about adding tags for replacement terms into OBO format. Jane: We are still not there yet. We are awaiting the release of OBO-Edit 1.1.
We'll contact mike when that happens.

d. Reference Genomes

Not much to say this week.
There is about to be an e-mail about the new April target genes (Monday).


f. User Advocacy

We are trying to merge the front pages of AmiGO and the GO website but it is proving difficult as they evolve independently. We are looking at mockups.

Possible go-help improvements:
Currently Mike is the moderator and eliminates a lot of spam from the list. In addition there is one person on duty each fortnight who makes sure that all mails are answered and forwards them to groups that might be best placed to answer them. Jane and Eurie ask if it would be good if all mails to go help went to a single gmail account that we could all access that the person on duty also did the spam filtering. Mike says this would not be a great idea as the spam filtering is quite subtle sometimes and he is very well practised at it and may as well go on doing it. Also there is no reason to burden the duty person with this as he is happy doing it.

The group is considering other possible improvements and will get back to us on that.


i. MGED10

Should we send someone to MGED10? (Australia - September)

Cons: We'd have to pay, and it's expensive and far away.

Pros: The microarray people are our largest user group.

Mike: MGED is changing from being mostly for tool developers to being more academic.

Midori: If we do go we will not do a general user meeting but could do something more targeted.

Jen: If we do go then instead of getting people to do academic talks couldn't we do a training day to teach microarray users how to use GO? We could do this jointly with one of the major tool developers. If we are paying for someone to go then Fiona McCarthy might be a good candidate as she is from Australia and goes there regularly anyway.

Action: No decision was made on whether to send anyone.
Jen will ask Fiona if she would eb interested in principle.


Institute of Animal Health (IAH): oxford. They have asked for a GO course.
Evelyn Camon did her PhD there so is interested. She has suggested a double act with Jennifer.

GOdatabase link

The godatabase.org link is changing so please don't use it any more. There will be a geneontology.org version coming out and we should use that instead.

Also NCBI should be changing their link to godatabase next week.

3. Meetings

Jane: Will be attending a plant ontology development meeting in Ghent.

Jane: Went to Manchester to see Robert Stevens for a conference on general ontology development stuff. 2 days long. Jane gave a talk.

She met some people with an animal behavior ontology that is not species specific and is well thought out with good ontology domain experts. She has suggested that we should work with them on the behavior terms. They'll be in contact.


4. Collaborations and Interactions, active and new

Jane is working with Dietrich Rebholz-Schuhmann at the EBI textmining group and Nicky Mulder of InterPro on mapping tool for ontology alignment. Currently too much time spent on making mappings so we need a tool. Deitrich's group is going to make one in the next couple of months and Jane is talking to them help with planning its design.


5. Operations and Procedures

i. Webex and skype communication.

The GO Editorial office and GOA are now all up and running with skype and finding it very useful indeed. We would like to encourage other groups to start using this facility to speed up discussion of GO issues. GOA are finding it particularly helpful for discussing annotation issues with remote curators. If anybody needs help to get started please get in touch.


6. Staffing and Personnel

7. Budget issues

8. Publications

i. OBO-Edit publication.

The applications note on OBO-Edit has been accepted. It is in "Bioinformatics" There should be an e-version any time. Congratulations to all involved!


9. Other Current Topics and Concerns