Uploaded image for project: 'Bug Repository'
  1. Bug Repository
  2. BUG-99

Hughes lab server missing data files, chokes IGB

    Details

    • Type: Bug
    • Status: Closed (View Workflow)
    • Priority: Major
    • Resolution: Fixed
    • Labels:
      None
    • Environment:

      IGB 6.5

      Description

      Some of the Hughes lab genomes on their QL site don't have the required files, e.g.,

      [java] WARNING: URL http://hugheslab.ccbr.utoronto.ca/igb/M_smegmatis_mc2-155_March_2008/mod_chromInfo.txt is not reachable, and is not cached!
      [java] Apr 14, 2011 6:33:33 PM com.affymetrix.genometryImpl.quickload.QuickLoadServerModel getInputStream
      [java] FINE: Load URL: http://hugheslab.ccbr.utoronto.ca/igb/M_smegmatis_mc2-155_March_2008/mod_chromInfo.txt

      IGB should alert the user when a QL site can't be used because it is missing critical files.

      Currently, it gives no warning - it just shows nothing when the site's contents.txt file is missing.

      IGB should instead open a dialog that says:

      "Warning: QuickLoad Data Source [NAME] does not contain required contents.txt file for requested genome version [GENOME VERSION]."

      where NAME is the name being shown in the Data Sources and Data Sets table.

        Attachments

          Issue Links

            Activity

            Hide
            anuj Anuj Puram (Inactive) added a comment -

            Passed v9908 on Mac

            Show
            anuj Anuj Puram (Inactive) added a comment - Passed v9908 on Mac
            Hide
            kkubiak Katharine Kubiak (Inactive) added a comment -

            Passed v9908 Windows

            Show
            kkubiak Katharine Kubiak (Inactive) added a comment - Passed v9908 Windows
            Hide
            hiralv Hiral Vora (Inactive) added a comment -

            Done is 8044.

            Show
            hiralv Hiral Vora (Inactive) added a comment - Done is 8044.
            Hide
            ann.loraine Ann Loraine added a comment -

            Hughes lab QL site contents.txt includes genomes (e.g., M smegmatis) for which it contains no mod_chromInfo.txt file.

            IGB adds M smegatis to the species and genome menus in the Data Access Panel.

            However, if the user chooses that genome, they see the Hughes Lab data source appear in the Data Access Panel, but no chromosomes appear in the Current Sequence Table.

            The problem is that IGB appears broken to users.

            I think that we should continue with the current scheme of adding genomes and species based on what QL sites advertise in their contents.txt files, unless there is a way to validate each one on startup. (Maybe there is??)

            Then, if the uses chooses a genome, all the data sources that claimed to have information about that genome should appear in the Data Access Panel, just as they do now.

            At that point, IGB queries the sites to get sequence information, lists of data sets, and so on.

            But...if the site doesn't have the necessary mod_chromInfo.txt file, then IGB should open a new dialog that says:

            "Warning: QuickLoad Data Source [Hughes Lab] does not contain required contents.txt file for requested genome version [M_smegmatis_mc2-155_March_2008]. IGB may not be able to display this genome."

            Our goal here is just to let users know that IGB is getting bad data and is not broken - it's the other site that is broken.

            Show
            ann.loraine Ann Loraine added a comment - Hughes lab QL site contents.txt includes genomes (e.g., M smegmatis) for which it contains no mod_chromInfo.txt file. IGB adds M smegatis to the species and genome menus in the Data Access Panel. However, if the user chooses that genome, they see the Hughes Lab data source appear in the Data Access Panel, but no chromosomes appear in the Current Sequence Table. The problem is that IGB appears broken to users. I think that we should continue with the current scheme of adding genomes and species based on what QL sites advertise in their contents.txt files, unless there is a way to validate each one on startup. (Maybe there is??) Then, if the uses chooses a genome, all the data sources that claimed to have information about that genome should appear in the Data Access Panel, just as they do now. At that point, IGB queries the sites to get sequence information, lists of data sets, and so on. But...if the site doesn't have the necessary mod_chromInfo.txt file, then IGB should open a new dialog that says: "Warning: QuickLoad Data Source [Hughes Lab] does not contain required contents.txt file for requested genome version [M_smegmatis_mc2-155_March_2008] . IGB may not be able to display this genome." Our goal here is just to let users know that IGB is getting bad data and is not broken - it's the other site that is broken.
            Hide
            hiralv Hiral Vora (Inactive) added a comment -

            User might not be aware about servers. So user doesn't need to know internals of servers even if something is wrong.

            Show
            hiralv Hiral Vora (Inactive) added a comment - User might not be aware about servers. So user doesn't need to know internals of servers even if something is wrong.

              People

              • Assignee:
                Unassigned
                Reporter:
                ann.loraine Ann Loraine
              • Votes:
                0 Vote for this issue
                Watchers:
                0 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: