Antenna is a cloud pipeline for the identification of sgRNA reads in viral next generation sequencing data.
The antenna pipeline can be run using any WDL execution engine, such as Cromwell or miniwdl. The pipeline was been tested using miniwdl version 1.4.3. The miniwdl package can be found here.
To install the python package you will require python 3.8 or later and the required packages listed under REQUIREMENTS.txt
Running the antenna pipeline does not require any installation if a suitable WDL run engine is available. In order to run the antenna pipeline export the package from github and run the wdl/antanna.wdl file.
git clone https://github.com/broadinstitute/antenna.git
miniwdl run --verbose -i input.json ../../wdl/antenna.wdl
If you wish to install the antenna python package you can do so using the following commands.
git clone https://github.com/broadinstitute/antenna.git
pip install -e antenna/
An example test script for the pipeline is provided under testing/testwdl/run.sh. This script will execute the WDL pipeline using inputs in the input.json file. The demo run takes less than 10 minutes to complete.
To run this example, please edit the file testing/testwdl/input.json and replace [LOCATION] with the download location of the repository in your system.
The pipeline outputs are as follows:
| Output Name | Description |
|---|---|
| tagged_bam | a bam file with additional tags denoting the presence of TRS sequences in the softclipped regions |
| output_counts | a summary csv file with TRS and non-TRD containing reads counts in each region |
| distribution_report | a png file of plot of the distribution of sgRNA containing reads in different orientations |
To run the pipeline on your data call the wdl with a custom input.json file. For example with miniwdl you would execute:
miniwdl run --verbose -i input.json ../../wdl/antenna.wdl
The docker build script uses the latest main version in github for installation. After updating github, run:
cd docker
./build [version]
docker push quay.io/nbarkas_1/antenna:[version]
You will need to create a custom quya.io repository if you wish to customize the image.