CNV-Mag

A tool to visualize CNV Events in WGS Data. CNV-Mag is taking use of BAM and short variant VCF files to visualize CNV events in WGS data in simple and direct way. It can help you to verify if a CNV event is real or not. This tool is part of the Broad Clinical Labs’s CNV reporting workflow.

Background

Verifying CNV events can be challenging, especially for multi-megabase events. A common approach involves using IGV to inspect read depth, soft clips, and SNP Allele Fraction (AF). However, this becomes impractical for large events. CNV-Mag addresses these challenges by providing a streamlined tool for visualizing CNV events. It also integrates Samplot, an excellent package for structural variant visualization, as part of its workflow.

Workflow

CNV-Mag takes BAM and short variant VCF files to visualize CNV events in WGS data. The workflow is shown below: Input:

• BAM file and BAM index file
  • Aligned BAM file of your sample
  • HG001 BAM file preferred using the same pipeline as above
  • HG002 BAM file preferred using the same pipeline as above
• Short variant VCF file (Requires AF field in the INFO column)
  • Short variant VCF file of your sample
  • HG001 VCF file preferred using the same pipeline as above
  • HG002 VCF file preferred using the same pipeline as above
• CNV region (Bed or a list of Strings)
  • Can be a single CNV region or multiple CNV regions
• Reference genome (hg19 or hg38)

There are three main workflows in CNV-Mag:

• Mag-SNP
  • Visualize all the PASS SNP AF of the CNV region.
• Mag-Depth
  • Visualize the read depth of the CNV region.
• Samplot
  • Visualize the discordant reads and split-read signals of the CNV region.

Running CNV-Mag on Terra

You can run CNV-Mag on Terra. The workflow is available at https://dockstore.org/workflows/github.com/broadinstitute/CNV-Mag/CNV-Mag:dev?tab=info. You can deploy the workflow to your workspace and run it on Terra.

Runtime specification

CNV-Mag is part of Broad Clinical Lab CNV reporting workflow. The run cost is less than $2 per run.