Run GVS on Verily Workbench

Running GVS on Verily Workbench

Prior reading: Workflows in Verily Workbench: Cromwell, dsub, and Nextflow


Introduction

Genomic Variant Store (GVS) is a WDL-based workflow developed by the Broad Institute. This tutorial shows you how to run GVS in your own workspace.

Step-by-step instructions

1. Create a Cloud Storage bucket to hold WDL files

Workbench currently requires that the WDL file(s) for your workflow is in a bucket. For this example, we create a new bucket.

  1. In the Resources tab, click + New resource and then New Cloud Storage bucket.
  2. Create a name for the bucket. In this example, we'll name it workflows_bucket. Click Create bucket.
Screenshot showing dialog for adding a Cloud Storage bucket.

2. Create a BigQuery dataset

The GVS workflow requires a BigQuery dataset to exist that it can read and write to. For this example, we'll create a new BigQuery dataset.

  1. In Resources, click + New resource and then New BigQuery dataset.
  2. Create a name for the dataset. In this example, we'll name it gvs_1. Click Create dataset.
Screenshot showing dialog for adding a BigQuery dataset.

3. Get the WDLs into the bucket

The WDLs used are available on GitHub here. To add them to the workflows_bucket created in step 1, run:

git clone https://github.com/verily-src/workbench-examples.git .
cd workbench-examples/cromwell_setup/gvs_wdls/known_good_wdls
gsutil cp *.wdl $BUCKET_NAME

4. Add the workflow

Navigate to the Workflows tab and click + Add workflow (or Add your first workflow if you haven't created any workflows yet). Add the WDL named GvsJointVariantCalling.wdl.

5. Create a new job

Click on the + New job button. Navigate to the next Prepare inputs page.

6. Enter the inputs

Enter in the following values:

Input Key Value Example
GvsJointVariantCalling.call_set_identifier Any string for this callset. "my_call_set_1"
GvsJointVariantCalling.dataset_name The dataset created in step 2. "gvs_1"
GvsJointVariantCalling.external_sample_names The list of sample names. ["2013050218", "2013050219"]
GvsJointVariantCalling.input_vcf_indexes The list of Google Cloud Storage locations pointing to the VCF index files of each sample. ["gs://genomics-public-data/ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20130502/ALL.chr18.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz.tbi", "gs://genomics-public-data/ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20130502/ALL.chr19.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz.tbi"]
GvsJointVariantCalling.input_vcfs The list of Google Cloud Storage locations pointing to the VCF files of each sample. ["gs://genomics-public-data/ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20130502/ALL.chr18.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz", "gs://genomics-public-data/ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20130502/ALL.chr19.phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz"]
GvsJointVariantCalling.project_id The Google Cloud Project ID of the workspace. "YOUR_PROJECT_ID"

7. Monitor the workflow

In the Workflows tab, click the Job status sub-tab to monitor jobs as they run and complete.

Diagram showing dialog for monitoring a workflow.

8. Get outputs

Once the workflow completes, browse the workspace bucket and navigate to the task containing the sharded VCF outputs.

Last Modified: 8 November 2024