Format for uploading an intervals file?

• 

  Adelaide Rhodes

  • Edited June 19, 2019 15:45

  Hi Lydia -

  This would be an error with the data model file that you are uploading to the workspace.

  Terra is looking for a header in the first column: 

  entity:workspace_id

  You can put whatever you want after the word "entity:"

  The first column header always needs to fit this format.

  Name the other columns to match the input parameters for your json.

  For example "bedfile" or "input_bed" for the column where you are listing your bed files.

  Here is an example I created yesterday to input bam files into IGV.

  entity:igv_id	bam	bai
  igvtest1	gs://terra-featured-workspaces/NA12878.chr20.sample.bam	gs://terra-featured-workspaces/NA12878.chr20.sample.bam.bai
  igvtest2	gs://terra-featured-workspaces/IGV_Demo/NA12878.wgs.20.bam	gs://terra-featured-workspaces/IGV_Demo/NA12878.wgs.20.bam.bai

   

  I called the file IGV.tsv and uploaded into the workspace.

  Let me know if you have more questions.

   

  Here is a helpful article on this topic:

   

  https://support.terra.bio/hc/en-us/articles/360026521831

  0
• 

  Brad Taylor

  • June 19, 2019 16:26

  Hello,

  If you are looking to upload a single interval list for use with your exomes, you can upload it to your workspace files by selecting "Files" under the data tab and clicking the plus icon.

  

   

  When configuring a workflow to run on your data, you can pick this file directly by clicking on the small folder icon.

  

  Each workflow you run will then use your bed file. This should be all you need to do to start running your analyses.

   

  A more advanced scenario:

  Workspace Data is useful if you want to set constants, or give your intervals file a useful name. Many of our featured workspaces make use of this. Here is an example workspace with data and methods for SNP / Indel calling

  https://app.terra.bio/#workspaces/help-gatk/Germline-SNPs-Indels-GATK4-hg38/data

  

  You can declare individual values, supplying a name and inputting the bucket path for your file (it starts with gs://). You can also upload them in bulk via a TSV. I think this is what you were trying to do, and getting errors because the system was expecting a TSV and seeing a bed file. If you would like to use to do this, I recommend downloading the TSV from the example workspace and looking at that.

   

  I hope this is helpful!
  
  

  Brad

  0
• 

  lck

  • June 19, 2019 16:34

  Thanks so much, Adelaide.

  I tried to follow your instructions, but still got the error: First column header should start with "workspace"

  So, I created a file like this:

  workspace:interval_id          intervals
  my_intervals                        gs://mybucket/itervals.bed

  Now it's at least loading without an error -- but I'm not sure if this is correct, since you'd said the file should start with "entity."

  0
• 

  Adelaide Rhodes

  • June 19, 2019 16:40

  It does not have to be entity.  It can either be entity or workspace.

  0
• 

  lck

  • June 19, 2019 16:58

  Thank you both -- I have it working now!

  0

Please sign in to leave a comment.