Talk:GMOD Evo Hackathon Proposal

From GMOD
Jump to: navigation, search

Meetings

We had several meetings while creating the proposal. The agendas and/or minutes for some of those meetings are here.

2010/02/12 Conference Call, February 12, 2010 2pm EST

The minutes taken by User:RobertBuels, any inaccuracies or omissions are his.

Agenda

Purpose of the hackathon

The hackathon was proposed to bring more support for evolutionary biology into the GMOD toolkit. We should thinks of what current needs are not being addresses or adequately addressed and take care to keep the focus on this topic. Some ideas I have are:

  1. Better support for comparative genomics (mainly MSA but also other types) data and data visualization. For example, bringing GBrowse_syn up to the same standard as GBrowse, improving support for the data processing and loading, useing an ontology such as the multiple alignment ontology for critical meta-data not captured bycurrent MSA formats.
  2. Others...

mckays 14:55, 12 February 2010 (UTC)

  • focus in on one or at most two implementation goals for the hackathon
  • identify which type of person we need to invite
  • start to develop a list of critical people to have there
  • develop some action items and assign them to people
    • drafting the proposal?
    • securing the venue?
      • NESCent has blocked out the second week of June or the last two weeks of July for us mckays 14:55, 12 February 2010 (UTC)
        • June timeframe overlaps with Arthropod Genomics (at this time, only Carson Holt is going from GMOD); nothing overlaps yet with last two weeks of July) - see Calendar for what else is going on.

Attendance

  • Sheldon McKay
  • Hilmar Lapp
  • Scott Cain
  • Robert Buels
  • Nicole Washington (2:25pm EST)

Sheldon - introduction

The purpose of the hackathon is to add more support for evolutionary biology in GMOD tools. I think hackathons are great, and GMOD will benefit from one.

The basic structure is that some number of people make a pitch for a thing to work on, and hope that people will nucleate around that and form working groups, and work for 5 days or however long.


Hilmar - mechanics of putting together a proposal

It falls into Nescent's whitepaper program, on the Nescent website, click on 'informatics' and the 'whitepaper program'. It's relatively informal, you put forward a proposed activity that convinces our directors to throw money at it. For hackathons, the amount of money is relatively significant, from $28-40K, depending on how many participants. The budget is significant, so you want to work out a relatively coherent argument as to why the activity itself is needed. How you expect to conduct it (which is not a big question in this case, we have done them before), and then what you expect to come out of it. What are the holes that you expect to be filling with that activity.

The proposal gets reviewed by the governing body of Nescent (currently called the operations committee, but that name could change soon).

When that is clear, the next step is to invite the key people and make sure you have those on board for the dates.

It's good to keep in mind that the whole process takes 4-5 months, allow a couple of weeks for writing the submission, submitting it, etc. Couple weeks for the invitations to go out and get responses. Couple weeks for an 'open call' for participants and responses.

Ideally, all the participants and invitations are known 2 months in advance. It's possible to violate that to some extent, but you don't want to stretch it too much.


Discussion

SM: For dates, I like the second week of June.

SC: I also am invited sometimes to arthropod genomics

HL: FYI, in July the ISMB conference ends on July 14. A significant fraction of the people would be going to ISMB, we need to be careful of that. Probably makes the week of July 19 not such a good candidate. The really oppressive heat here starts in July and ends mid-August, when it's oppressively hot here.

SM: I really want to try to make June work for the reasons you laid out. I have some other stuff in early July, then ISMB and the plant biology meeting.

NW: early June would also work for me.

Group decided to shoot for week of June 7-11.


SM: We need to think about the kind of people we want to invite, we need people that can actually sit down and implement working products at this session.

HL: I would suggest with drafting the proposal whitepaper, the key people would probably naturally fall out of that. The first step would be to start fleshing out some ideas for what good focus areas would be. Unless you think we need those key people to put us in a position to flesh those out in the first place.

SM: Whitepapers from previous accepted hackathons, are those available for us to see?

HL: Yes, there are some linked from their respective home pages. HL TODO: provide links to the accepted whitepapers

HL: I wouldn't go too overboard with the whitepaper, what is more critical than exhausting it there is to keep pushing forward within the time-frame and nailing down key points.

SM: Given our timeline, I'd say we should probably be communicating offline quite a bit and iterating very fast on the whitepaper. For the mechanics, I think it would be good to set up a google doc and a google group mailing list


Personal Wishlists

SM: I think it would be good if we each stated the personal agendas we have in participating in this hackathon.

SM: My agenda is improving comparative genomics support for things that are in the GMOD toolkit. Bringing the synteny browser up to a standard similar to GBrowse is near and dear to my heart.

RB: My personal agenda is good software engineering. I want to make sure that things are put together well.

SC: Anything having to do with GMOD, I'm there.

HL: Chado natural diversity module, handling taxonomies well, handling phylogenies. Subgroup sizes of 4 to 5 in working groups are by far the best.


Action Items

SM: set up google doc for the whitepaper

SM: set up organizer mailing list for organizing committee, and one for the participants

SM: send out a Doodle poll for the next conference call

HL/SM: find links to past accepted proposals

2010/02/22, Second conference call, 5PM EST

Agenda

Please add/change items.

Attendance

Sheldon McKay, Hilmar Lapp, Scott Cain, Nicole Washington

Proposal ideas

  • we can craft the participants with ideas of topics, but the reciprocal is true. up to participants to decide. organizers are also participants.
  • scott - proposal is more of the "these are the things we 'might' do", rather than exactly what we do. its an overarching guide.
  • hilmar - trying to decide what to accomplish, and gather desires; but what is actually accomplished, is executed by participants. the final thing is usually close to what is produced; the white paper set the tone.
  • don't need really polished white paper. but a good outline. this could probably be done quickly.
  • scott - need two pronged approach: write white paper & identify topics/people, do this in parallel.
  • scott - we can do the white paper in the next couple of weeks, and then start letting people know.
  • come up with skeleton of proposal, and then dole out parts. sheldon/hilmar work on first draft. and then rest comment on during the rest of the week. hilmar wants to put it on the wiki, so long as people's names aren't mentioned. could do google doc if privacy is needed. decision, this will be moved to wiki. will use old wish list as jumping off point.
Action item - Sheldon to create new wiki page
Action item - Sheldon/Hilmar to write skeleton of proposal.
Action item - Sheldon/Hilmar to solicit input from Scott/Nicole/Robert
  • dates recommended are second week of June (7-11)
  • size ~20 persons. not more than 30.
  • may seek some extra funding for additional participants.
  • will this be open call or invite only? would be good for open call for people to apply...
  • next call in two weeks.

2010/03/17 Third conference call, 2PM EST

Agenda

Please add/change items.

  • Focused on getting the proposal written.
    • clarify metadata issues
  • Dole out responsibilities

Attendance

  • NLWashington
  • ScottCain
  • RobertBuels

Discussion

  • We are running out of time, we need to get this proposal submitted.
  • We should get Dave Clements to look at the proposal. TODO: Rob will get him to do so.
  • Let's set a goal of getting the proposal submitted by the end of next week (Mar. 26)
  • Scott volunteers to chair next week's meeting.
  • Specific editing action items
    • Nicole: add more about alignment metadata
    • Dave Clements: check on natural diversity module and its relation to evolutionary phenotype data
    • Hilmar: general edits
    • Rob: general edits


2010/05/05 conference call, 5PM EST

Attendance

  • Dave Clements
  • Nicole Washington
  • Rob Buels
  • Hilmar Lapp

Last Meeting's Action Items

  • Nicole: add more about alignment metadata
  • Dave Clements: check on natural diversity module and its relation to evolutionary phenotype data
  • Hilmar: general edits
  • Rob: general edits

Discussion

Discussed possible inclusion of atlas-based interfaces for phenotype, expression, and fate map data. This came out of a recent NESCent working group. This would be attractive to support evo-devo work.

Have Motivation section be overview. Details will be in the Specific Objectives sections. NESCent doesn't actually care about structure. The document should just get its point across clearly, both to NESCent reviewers, and to the GMOD community.

Issues that affect everyone and are not specific to evolutionary biology should not be in the proposal. If such are issues are needed by all/most evolutionary biologists, then they will emerge at the hackathon. Dropped schema evolution objective.

Need to specify budget and some recommended requirements. Budget largely driven by number of participants, and that is largely driven by the number of specific objectives being proposed. We should get good faith intent to participate commitments from anyone who we think is vital to have at the event.

Should we investigate integration between phylogeny, organism, and natural diversity module? If so, we should identify any pain points and address them.

Visualizing phylogenetic data in GBrowse_syn came up. We decided that NGS data in GBrowse_syn was of vital interest to the evo community, based on recent working group at NESCent that Dave and Hilmar contributed to. Getting GBrowse_syn into the GBrowse 2.0 infrastructure is a key item. (But I wasn't clear on if this should happen at the Hackathon, or before it.) We determined that we should not do any work that rests on top of the GBrowse 1.x platform.

Rob suggested "soliciting gripes" as a good way to identify work items and then what could be done about them.

Action Items

  • Dave will do first pass on initial restructuring of proposal and then pass it to Nicole. -- Done 2010/04/06, but not as polished or full as it could have been.
  • Dave will create Google Docs/Google Groups pages for budget and participants lists. -- Done 2010/04/05.
  • Nicole will write up details for alignment metadata objective.
  • Nicole and Dave will jointly work on comparative phenotypes in Chado objective.
  • Rob will talk to Naama about her experience with the Chado Phylogeny module.
  • Dave will contact a key GMOD contributor to find level of interest in participating.
  • Dave will writeup details for GBrowse_syn in Specific Objectives and then send it to Sheldon. -- Written up, sort of; Need to finish/discuss with Sheldon. 2010/04/06