bihealth / seasnap-pipeline-browser

Docker image build instructions for the sea-snap pipeline browser

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Pipeline browser for sea-snap

This docker image contains the pipeline browser for the sea-snap pipeline. To use it, do the following:

  1. Contact Matthias or Mikko to link your SODAR project to a KIOSC project.

  2. On SODAR, create a landing zone (LZ) for the data. Record the IRODS path to the LZ. For the example pipeline data in sea-snap, this is

    /sodarZone/projects/1e/1e526322-50b8-47dd-a419-759d03d19a0b/landing_zones/weinerj@CHARITE/seasnap_test_data/monocytes/20210812_165144
    
  3. On the HPC, create a .tar.gz file containing the pipeline including the yaml config file. If you used default settings, this amounts to running

     tar hczf DE_pipeline.tar.gz DE_config.yaml DE
    
  4. Copy the .tar.gz file to the LZ, for example with

     iinit
     iput DE_pipeline.tar.gz /sodarZone/projects/1e/1e526322-50b8-47dd-a419-759d03d19a0b/landing_zones/weinerj@CHARITE/seasnap_test_data/monocytes/20210812_165144
    
  5. Create a ticket (token) for anonymous access to the IRODS LZ:

     iticket create read /sodarZone/projects/1e/1e526322-50b8-47dd-a419-759d03d19a0b/landing_zones/weinerj@CHARITE/seasnap_test_data/monocytes/20210812_165144
    
  6. On the corresponding KIOSC site (the link is e.g. in SODAR under "This project on other sites", or simply go to kiosc.bihealth.org) create a new container ("Containers" -> "Create container"). Steps below describe how you should fill out the "Create container" form.

  7. Under "repository", enter ghcr.io/bihealth/seapiper. Under "Tag", enter 0.1.1 (current version of the container). Under Container port, enter 8080 (the port at which container exposes the pipeline browser). Note: it happened to me more than once that I have inadvertently changed the port number by using the laptop touchpad. This results in a "server error" without any sensible error message, so make sure that the port number is correct!

  8. Now you need to configure the variables under "Environment". This entry is in JSON format and must include the information for the container how to access your data. Here is a template. The actual tocken from the iticket step above was replaced here by XXX, but you need to enter the real token of course.

     {
       "datasets":[
        {
          "name":"Example data set",
          "archive":"DE_pipeline.tar.gz",
          "config":"DE_config.yaml"
        }
       ],
       "IRODS_PATH":"/sodarZone/projects/1e/1e526322-50b8-47dd-a419-759d03d19a0b/landing_zones/weinerj@CHARITE/seasnap_test_data/monocytes/20210812_165144",
       "DAVRODS_SERVER":"davrods-anonymous.sodar.cubi.bihealth.org",
       "IRODS_TOKEN":"XXX",
       "TITLE":"Sea-snap example project"
     }
    

    The "datasets" entry in the JSON format above allows you to define multiple archives (workflow results) which you can than view with a single interface.

  9. Enter IRODS_TOKEN under "Environment secret keys" such that the token cannot be viewed by other users of the Kiosc site. When you edit the container again, the actual value of the token will be shown as <masked>

  10. Click on "Create", and the start the container. Presto! Go grab a coffee, because starting the container will take a few minutes. Had your coffee? The container should now be available. Or not, and in such a case, contact me and we will figure out what went wrong.

About

Docker image build instructions for the sea-snap pipeline browser


Languages

Language:R 61.7%Language:Dockerfile 34.6%Language:Shell 3.8%