Skip to content

Latest commit

 

History

History

3_quality_control

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 

Step 3: Quality Control

This folder contains two Jupyter notebooks outlining five quality control steps. Between the first and second notebooks, you will need to manually curate the experimental metadata to identify which samples are biological/technical replicates.

To run these notebooks, you can either install Jupyter Notebook and install the dependencies in requirements.txt, or use a pre-existing docker container.

Running the docker container

  1. Install Docker
  2. Run the following code in terminal
docker run -p 8888:8888 avsastry/modulome-workflow:v1.0
  1. Select the third link in terminal (starts with 127.0.0.1)
  2. Navigate to the 3_quality_control folder and open the notebooks. Any changes made are only saved in the container, and not on your local computer.

Mounting local files in the Docker container

If you want to edit and save files in the docker container, replace the above command with the following:

docker run -p 8888:8888 -v <target-path>:/home/jovyan/work avsastry/modulome-workflow:v1.0

The above commands will mount the files in your local <target-path> to /home/jovyan/work in the Docker container. These files are editable, and changes made in the docker container will be reflected on your own machine. Only subfolders and files within <target-path> can be accessed by the container, so it is recommended to input the root folder of this repository as the <target-path>.

For more options, such as changing the default username, changing the port, or granting root access, see the Jupyter Docker Stacks Feature page