TR110531

Minutes

  1. Release 1.0 software update (Naim)
    1. Ray will have increaded availability to work on TR. First task is to modify the UI to support multiple reconciliations.
  2. Documentation for release 1.0 at TR_Documents (All)
    1. There will be separate, independent documentation for the available datasets and for the software. The software documentation will include the description of the pipelines.
    2. Jim will create a powerpoint presentation to be used as a storyboard for the tutorial.
    3. Cecile points out that the examples will have to be database specific, which might be problematic for providing online example for the search fields. Naim to investigate if interface can support different examples for different datasets.
  3. GSoC project description (Jamie/Daniel)
    1. Daniel is working on a skeleton Bioperl software implementation and hope to have something by the end of the week.
    2. Rutger Vos and Chris Zmasek should be involved and contacted for questions. Also, questions can be sent to the group mailing list.
    3. Daniel to copy the working group on his weekly updates.
  4. Status of modificaions to allow selection among multiple pipelines and studies
    1. In process. Open question is how to identify different datasets in the database.
  5. Bowers data analysis
    1. Status of RaXML-NOTUNG pipeline
      1. Sheldon to coordinate with John Kerry
  6. 1KP data and pipeline
    1. Status of PrimeGSR pipeline on HPC
      1. Sheldon to test pipeline on CSHL cluster before porting to TACC. Might require refactoring of code.
      2. Discussion on how to handle polytomies in the gene trees resulting from highest support value <50%. Would be ok to keep polytomies but creates problem if speciation and duplication nodes are collapsed. Preferred solution: display the best tree with support values for branches with >50% support.
    2. Status of gene family circumscription and alignment steps
      1. Pilot (91 species): 90% done.
      2. Mike Barker to work on remaining taxa, expected time to completion: ~2 weeks. Then the analysis of families can start.