DRAGEN Reports is a docker image that provides tools for generating rich, interactive and self-contained HTML reports from DRAGEN's output files. These reports combine data from QC, trimming, mapping, variant and other DRAGEN modules to create a comprehensive summary of a multi-sample workflow, as well as more detailed reports for individual samples.
The DRAGEN Reports tools are provided as a Docker image. To use it, you will need the following:
An x64 processor
Docker installed and running
In addition, while DRAGEN Reports can accept files output by any DRAGEN process (such as from a local DRAGEN box, DRAGEN equiped sequencing instrument, or the Cloud), it can only work with local files. For how to access your data and transfer it to your local environment, please refer to the user documentation for the appropriate service or instrument.
Within the DRAGEN Reports docker image are multiple executables for generating different types of report files
dragen-reports - Generates reports for a single DRAGEN workflow or pipeline and it's samples
dragen-summary-reports - Generate reports for a NovaSeqX run, containing multiple containing multiple DRAGEN workflows
The dragen-reports tool produces both workflow- and sample-level reports. A workflow-level report combines data from multiple samples analyzed by DRAGEN in the same way in order to faciliate comparssions across a batch of samples. Users can navigate to a report for a specific sample by clicking on the link in the Sample column of most workflow report tables.
Sample-level reports are also generated to allowed for more detailed analysis of individual samples. For example, under the DRAGEN-FastQC tab in a workflow report we are only able to plot the mean base quality, while for individual samples we can display a box-and-whisker plot for each position, as shown below.
To navigate back from a sample report to the multi-sample summary, users can click on the "Back to Pipeline" button located in the upper-left corner of the report.
The following command-line options are available from DRAGEN Reports
In addition, the following Docker options are recommended when calling DRAGEN Reports via Docker:
The dragen-reports tool can output reports with different contents for different DRAGEN workflows, tailoring the resulting reports to create only the most relevant plots and tables. To utilize this feature, users can use the -m / --manifest option to specify a Manifest JSON file to specify the desired output reports. These files function as recipes for different reports - specifying which files to look for, which data to process, and which plots to draw. These files are all available in the DRAGEN Reports docker under the /opt/dragen-reports/manifests directory.
Below is a sample Bash script for running the dragen-reports tool from the DRAGEN Reports docker image, with a line-by-line explanation
Call to Docker to mount a new container and execute a command in it
Specify the platform being run. In this case I am running on a Mac, so I specify that I need an image that can run on an amd64 processor
Mount the input data from our local machine to the Docker container
Mount the desired output location from our local machine to the Docker container
Docker remove argument, tellind Docker to unmount and delete the container when the process is finished
Specify that Docker image to be run, in this case dragen_reports v4.3.0
Call to execute the dragen-reports command-line tool within the container
--force option, to enable over-writing of any pre-existing output files
--directory option specifying the input data directory mount on Line #3
--output option specifying the report file to output
--manifest option to specify the manifest file to use - in this case we are requesting a WGS Germline report
The dragen-summary-reports tool produces reports for DRAGEN analysis results from a NovaSeqX instrument. These reports summarize which DRAGEN workflows were run, which version of DRAGEN was used, which sample project they belonged to (if any), and how many samples and errors there were. If a workflow-level report is available for a given workflow, such as one generated by dragen-reports, that report will be linked from the Workflow column. In addition sub-reports are created for any sample projects, containing the subset of data for that specific project, and linked from the Sample Project column.
The following command-line options are available from DRAGEN Summary Reports
In addition, the following Docker options are recommended when calling DRAGEN Summary Reports via Docker:
Call to Docker to mount a new container and execute a command in it
Specify the platform being run. In this case I am running on a Mac, so I specify that I need an image that can run on an amd64 processor
Mount the input data and output location from our local machine to the Docker container
Docker remove argument, tellind Docker to unmount and delete the container when the process is finished
Specify that Docker image to be run, in this case dragen_reports v4.3.0
Call to execute the dragen-summary-reports command-line tool within the container
--force option, to enable over-writing of any pre-existing output files
--directory option specifying the input data directory mount on Line #3
--output option specifying the report file to output
NOTE Due to the fact that dragen-summary-reports makes extensive use of symbolic links to prevent the duplication of large report files, the --directory and --output arguments must point to the same mounted volume
Option | Argument | Description |
---|---|---|
Option | Argument | Description |
---|---|---|
Manifest | Name | Pipeline | Description |
---|---|---|---|
Option | Argument | Description |
---|---|---|
Option | Argument | Description |
---|---|---|
-V, --version
Output the version string of the DRAGEN-Reports executable
-v, --verbose
Emit additional debug information whiile generating the report
-m, --manifest
Filepath
Specify a manifest file describing the report to be written
-n, --run-name
String
Manually set the run-name in the output report
-d, --directory
Comma-separated list
DRAGEN output directories to generate a report from
-o, --output
Filepath
Generate output report to the specified location
-s, --samples
Comma-separated list
Generate only individual reports for the specified sample(s)
-S, --sample-output
Filepath
Generate sample report to the specified location
-f, --force
Force overwriting of any pre-existing report files
-T, --timestamp
String
Manually set the timestamps in the output reports
-h, --help
Display this help command
-v, --volume
String
Bind mount a volume
--rm
Automatically remove the container when it exits
--platform
String
Set platform if server is multi-platform capable
germline_wgs.json
Germline Report
DragenGermline
Default manifest targetting WGS workflows and supporting most DRAGEN variant callers
somatic_wgs.json
Somatic Report
DragenSomatic
Manifest for somatic tumor-only analysis
germline_enrichment.json
Enrichment Report
DragenEnrichment
Manifest for whole exome or germline gene panel analysis
somatic_enrichment.json
Enrichment Report
DragenEnrichment
Manifest for somatic exomes or somatic gene panel analysis
rna.json
RNA Report
DragenRna
Manifest for RNA-Seq & RNA Quantification workflows
methylation.json
Methylation Report
DragenMethylation
Manifest for the DRAGEN Methylation workflow
demux.json
Demultiplex Report
Demux
Manifest for DRAGEN demultiplex processes
bcl_convert.json
BCL Convert Report
BCLConvert
Manifest for BCLConvert and library QC workflows
-V, --version
Output the version string of the DRAGEN-Reports executable
-v, --verbose
Emit additional debug information whiile generating the report written
-n, --run-name
String
Manually set the run-name in the output report
-d, --directory
Directory path
DRAGEN output directories to generate a report from
-o, --output
Filepath
Generate output report to the specified location
-f, --force
Force overwriting of any pre-existing report files
-h, --help
Display this help command
-v, --volume
String
Bind mount a volume
--rm
Automatically remove the container when it exits
--platform
String
Set platform if server is multi-platform capable