• Jump to Call-in/Webex information
• Jump to Notes section

Call-in/WebEx Information

• TR_WebEx - this meeting will include visuals, so WebEx Video will be utilized

Action items

• AI: Former user (Deleted) to mockup an expert tree prune/selection w/ auto-complete to find taxa. Common taxa box should be incorporated. APWEB has an example of how to show users where their species/group of interested is placed on the tree. [incomplete]
• AI: Andrew needs to speak with Steve Goff to see if next-gen assembly is on radar and potential actions iPlant can do to aide efforts. [done, 15-Mar-2010]

Agenda

1. Brief review
   • Update re conversation w/ Steve G.
2. Review Underlying Pipeline
   • Prototype Pre-Compute Workflow
   • Issues experienced by Sharon
3. Revisit original "proof-of-concept"

Open Questions

1. It was suggested that if the entire subfamily/cluster was not select (so particular members were checked in the checkbox interface, but not all) we might want to use some visual representation of this.
   • How do we represent subfamilies or families not checked by the user?
   • Answer: ?

Notes

(Previously schedule meeting on March 29, 2010 had been rescheduled to April 5, 2010)

Attendees: Todd Vision, Sheldon McKay, Adam Kubach, Jerry Lu, Nicole Hopkins, Andrew Lenards

• Discussion regarding the Prototype Pre-Compute Workflow
  • In the optional (grey-shaded steps) portion, there may be some steps missing between "Acquire nucleotide data" and "create gene clusters".
  • Clarify "Create Gene Clusters" - this forming clusters based on sequence similarity between genes in different species (say, maize and arabidopsis).
  • For now, we are assuming that we have AA-guided alignments and are beginning from the start indicator.
  • Assuming the AA-guided alignment & large species tree
    • Issue: the tree requires formatting to be 5-character Swiss-Prot unique labeling.
      • Sheldon does not think this is a significant problem - easily solved with a lookup table.
  • Feed input into TreeBeST and get out gene trees with annotations in NHX format.
  • Would then be processed by PrIMETV... but Sharon is having issues with PrIMETV in her 64-bit environment
    • Jens group has been very responsive in answering questions for her.
  • Persist results & metadata
    • original idea was to use something like the Treefam Schema as a starting point
    • metadata is referring to the data-model required to link into the entry pts (search & blast)
• Sharon has contacted Dr. Li, author of TreeBeST regarding the 5-char, Swiss-Prot constraint. But the topic of doing reconciliation via TreeBeST was not broached. The project's mailing list was not contacted. Todd suggested contact Dr. Li with regard to contributing resources to improve the tool.
• Discussion about the original 'proof-of-concept'
  • Prime-GSR was mentioned.
    • Computationally intensive, slow - it has been optimized for correctness and not speed. May need to look at speeding it up, 2nd phase of reconciliation.
    • Key contribution is that it provided inherent confidence values on the reconciliation.
• Pre-computation pipeline helps satisfy Use Case #1
• AI: A draft of the data model to support dependent use cases will be prepared by Sheldon & Sharon for the next meeting.
• What the completed prototype would be proving as a 'proof-of-concept':
  • (a paraphrase of Todd Vision's statements)
    • The precomputed reconciliations with the proper supporting data model will provide an interface for researchers to ask questions. They'd be able to do things with the interface that can't currently be done; be able to ask questions of reconciled data. Todd wants to share the prototype with various groups outside of iPlant to get suggestions and help grow the functionality. Todd wants to show the value of the "Big Tree" to people who may only care about reconciliations.

Next meeting: April 19, 2010