Using the Docker image

A docker image of MARTi, which includes all of the required dependencies, is provided on Docker Hub. To get started, ensure that you have installed the Docker Engine on your system. Once installed, you can easily pull the MARTi image by running the following command:

docker pull nedpeel/marti

This will fetch the latest version of the MARTi image and make it available on your system for use. To run the MARTi Docker image, execute the following command:

docker run -i -t -p 3000:3000 -v /path/to/your/databases:/usr/databases -v /path/to/your/sequencing_data:/usr/reads -v /path/to/your/marti_output:/usr/output nedpeel/marti

This will start the container and present you with a prompt where you can enter your MARTi commands. For example, to run the MARTi Engine, use the following command:

marti -config <file> [options]

To exit the Docker container, simply type exit.

Notes:

In the docker run command, you need to map your database, data, and output directories to the Docker image. This is done with the -v option. In the above example, the databases and taxonomy directories on our local machine are in /path/to/your/databases and this appears in the Docker image as /usr/databases.
If you get an error from the docker command, it may be because you haven’t sudo’d it, or added your user to the docker group - see How can I use docker without sudo?

Test data

If you want to test the MARTi installation, you can use a set of sample reads, custom database, and taxonomy files available for download from here.

After downloading the files, unzip them and run the MARTi Docker image with the following command (replacing /path/to/marti_example/ with the actual path to the unzipped marti_example directory):

docker run -i -t -p 3000:3000 -v /path/to/marti_example/databases:/usr/databases -v /path/to/marti_example/reads:/usr/reads -v /path/to/marti_example/output:/usr/output -v /path/to/marti_example/config:/usr/config nedpeel/marti

Once the container is running, start the analysis with MARTi Engine running in the background with the following command:

marti -config /usr/config/marti_test.txt > marti_out.txt &

Launch the MARTi GUI by running marti_gui, open a browser and navigate to GUI’s port localhost:3000.

Persistent options file

The marti_engine_options.txt file is used by both the Engine and the GUI. Users can set paths to their taxonomy directory and define blast processes, making MARTi config generation easier (see Starting analysis for more information about creating a config file and defining Blast processes). This file is also used to set the read data and MARTi output locations for the GUI to monitor (see Download and installation). However, changes made to the options file within the Docker (~/marti_engine_options.txt) will be lost on exit. For a persistent options file, copy the file to your local machine, map the directory containing the file to the Docker image, and then run the Engine and GUI with the -options flag followed by the path to your options file. For example:

marti -config <file> -options /path/to/marti_engine_options.txt [other options]

marti_gui -options /path/to/marti_engine_options.txt

GUI only

To view previously analysed samples, the MARTi Docker image can be used to run only the GUI component of the tool. When running the Docker for this purpose, you only need to specify the MARTi output directory as a volume:

docker run -i -t -p 3000:3000 -v /path/to/marti_example/output:/usr/output nedpeel/marti

Once the container is running, launch the MARTi GUI by running marti_gui. To view the GUI, open a browser and navigate to GUI’s port localhost:3000.