Phylotastic/Schedule

From Evolutionary Interoperability and Outreach
(Redirected from PhylotasticSchedule)
Jump to navigation Jump to search

Phylotastic: ToL's Killer App, in Just 5 Days

Day 1

Session 1 (morning): User scenarios

The idea: The reason Phylotastic is so exciting is that it finally will facilitate a useful and highly automated flow of Tree-o-Life knowledge from producers to consumers. The key to creating this flow is to focus on what users need. Thus, after an initial overview on re-use of phylogenetic knowledge, we will begin with talks describing how an imaginary phylotastic will be used in scientific research.

How to participate

  • volunteer (alone, or with others) to develop and present a user scenario
  • ask questions to understand user needs

Detailed schedule (add a link to your presentation-in-progress here)

Session 2 (late morning): Brainstorm

The idea: In this session, the entire group will participate in a high-level design discussion. Now that we know what Phylotastic will be used for, what components and connections will work together to satisfy user needs? This is not the time for technical discussions. Technical discussions will be allowed to the extent that they inform high-level design choices.

How to participate

  • be ready to contribute all of your experience and knowledge to design decisions.
  • Notes on Google doc

Session 3 (early afternoon): Boot Camp

The idea: Many participants will benefit from a rapid introduction to key technologies (e.g., NeXML, TNRS, etc).

How to participate

  • request a boot camp on a particular technology
  • offer to give a boot camp on a particular technology.

Detailed schedule

Session 4 (late afternoon): Self-assembly

The idea: Participants will self-assemble into sub-groups to work on specific tasks. Presumably most task groups will focus on a particular component, however some groups may have a multi-component focus on documentation, testing, integration, etc.

How to participate

  • start thinking about what is the best use of your time and talent
  • propose an idea (called a "pitch") for a task group with specific deliverables
    • deliverables
    • approach
    • key challenges
    • strategic partners (how are you going to work with other groups)
  • ask critical questions (about the value of proposed deliverables, and about the manner of achieving them)

Detailed schedule

  • 3:15 pitches (a few minutes each) list of pitches
  • 3:40 self-assembly of groups
  • 4:30 re-pitches
  • 4:50 gap analysis and integration

Session 5 (after dinner): Self-disassembly

The idea: unwind, get to know fellow participants.

How to participate: Meet at Fullsteam Brewery at 7:30 pm. This is a short half mile walk from Duke Tower. Here's a map. On Monday nights, a local chef cooks up an Indian buffet from 5:00-8:00, if you want to grab food there.

Day 2

an interruption in Durham water service is expected for the afternoon. The leadership team is working to ensure that we have access to bathrooms during the afternoon

Morning work session 9:30 to 12:00

  • coffee breaks as needed
  • individual groups work on project goals, record progress on wiki
    • design decisions
    • challenges identified
    • links to any tangible outcomes such as code or documentation or figures
    • next steps
  • core group produces an API sketch

iPlant Tools and Services for Phylogenetic Analyses, Naim Matasci, 12:00 pm

  • lunch will be provided?

Reports 1:15

Groups consider API sketch 2:00

  • individual groups make notes on API sketch

Discussion: draft Phylotastic specification 3:30

Days 3 and 4

The People Want to Code (not talk)! Please just go straight into your task groups this morning!

  • coordinate with other groups
  • make yourself available to remote participants
  • coffee breaks as needed
  • lunch from 12:30 to 1:30

All participants meet 4:45 end-of-day discussion

Day 5

Morning session 9:30 to 12:30

  • coffee breaks as needed

Wrap-up session 1:30 to 3:00

final reports from subgroups

Branch length group

  • nice web page with various options
  • planned lightning talk iEvoBio (Brian taking the lead)
  • publication (Brian taking the lead)
  • phyloorchard database

TNRS

  • unified API
  • demo implementation that works
  • some stuff implemented but not exposed (e.g., classification)

TreeStore

  • triplestore with trees, pruning, etc.
  • extensions to CDAO

Architecture

  • js controller (Helena)
  • manuscript doc started
  • galaxy stuff; with plan to improve (Rutger)
    • possible galaxy server Netherlands ?
  • map-reduce uses (Rutger and Aaron)
  • phylomatic version 3 reads & writes nexml & cdao (Cam)

Shiny

  • reconcili-o-tastic works
  • Mesquite-o-tastic works

Notes from final discussion

  • Possible publications
    • applications of map-reduce in phyloinformatics - Rutger and Aaron
    • DateLife - the Brian O'Meara
    • TNRS - description of generalized API
    • hackathon - description of process and products (Arlin)
      • venues: Biol Letter; Methods in Ecol & Evol; PLoS Currents Tree-o-Life (quiescent?)
  • posters and talks at upcoming meetings
    • iEvoBio "full" talk (accepted) - 10 to 15 min by Karen or Arlin
    • live demo at iEvoBio (Mark H)
  • future meetings
    • SMBE 2013 (Chicago) call for satellite meetings
    • next year's iEvoBio
  • building out what we have done already
    • TNRS -- Guarav may be doing this for his thesis work
    • reconcile-tree case -- Christian wants to build client capacity into SDI
  • getting stuff on the dev server
    • DateLife is remote, but its R, so it should be easy to use (Brian)
    • TNRS also remote, but can be moved easily (Naim)
    • TreeStore stuff (triplestore and web server) are on dev server
    • shiny stuff is all local
    • Rutger's galaxy stuff is remote