OpenCRAVAT reports
OpenCRAVAT can output annotation results in various formats to support different ways of viewing results, and integration with downstream analysis.
Outputs are created by running the oc report command on an OpenCRAVAT sqlite database.
oc report job.sqlite -t tsv excel
The -t flag is required to select one or more reporters to run. Multiple formats can be generated in a single command. The exact output files created are listed in the terminal output:
$ oc report job.sqlite -t tsv excel
Generating tsv report... report created: job.variant.tsv
Generating excel report... report created: job.xlsx
Output files are created in the same directory as the sqlite database. Use -d to write to a different directory, and -n to set the output filename prefix.
Discovering and installing reporters
List reporters you have installed:
oc module ls -t reporter
Browse all available reporters in the store:
oc module ls -a -t reporter
Install a reporter:
oc module install reporter_module_name
oc report options
Option |
Description |
|---|---|
-t [FORMAT [FORMAT …]] |
Report type(s) to generate: |
-d OUTPUT_DIR |
Directory for output files (default: directory of the sqlite file) |
-n RUN_NAME |
Output filename prefix (default: input filename) |
-f FILTERPATH |
Path to a filter file |
-F FILTERNAME |
Name of a filter stored in the sqlite database |
–nogenelevelonvariantlevel |
Prevent gene-level results from being added to the variant-level output |
–separatesample |
Write each variant-sample pair on a separate line |
–module-option MODULE.KEY=VALUE |
Pass a configuration option to a specific reporter module,
e.g. |
-c CONFPATH |
Path to a configuration file |
–confs CONFS |
Configuration string |
Report formats
excel
Produces a .xlsx spreadsheet. This is the default reporter included with a base installation.
tsv
Produces a tab-separated values file per requested level. By default only the variant level is written (job.variant.tsv). To include other levels, use --module-option tsvreporter.pages=variant,gene (any combination of variant, gene, sample, mapping is accepted). Each requested level produces a separate file.
text
Writes all four levels (variant, gene, sample, mapping) to a single file by default.
csv
Produces a comma-separated values file. Accepts the same options as the TSV reporter.
vcf
Produces an annotated VCF file with OpenCRAVAT annotations added to the INFO field.
Note that if the input was not a VCF file, the output will lack fields such as zygosity or read depth. If you are annotating a VCF input and want a fully annotated VCF output, see Large VCF Annotation.
Building your own reporter
See the Reporter tutorial page for documentation on writing a custom reporter module.