Inferring demography and selection using an end-to-end deep learning approach, using dnadna

Flora Jay and Jean Cury, June 2022


How to run this activity

The activity on inference of selection using the machine learning approach packaged in the software dnadna will be run through Jupyter Notebooks. So, just as in the introductory tutorial on Python and Jupyter Notebooks by Hannes Svardal, you will first need to start a Jupyter Notebook on your AWS instance and connect to it.

Connect to the instance using SSH. Use “wpsg” as the username and the usual password. You can do that either through Guacamole in your web browser:

Or you can use SSH from your terminal:

Once connected through SSH:

  • Navigate into the tutorial directory: cd ~/workshop_materials/a14_dnadna/
  • Start a new screen session by typing: screen
  • Confirm with the “Return” key
  • Start the conda virtual environment: conda activate dnadna
  • Start the notebook server: jupyter notebook --no-browser --port=8889
  • The command blocks the terminal. That is normal. Keep it running. You can detach from the screen and get back to a functional terminal by typing “Ctrl + a, d” (first “Ctrl + a”, then “d”). (In case you wanna go back to the screen, type screen -r )
  • In your local browser, navigate to the web address: http://ec2-XXX-XXX-XXX-XXX.compute-1.amazonaws.com:8889 (replace XXX with your Amazon instance IP address
  • Then you should see the contents of the folder in which your instance is running.
  • Find the file called WorkshopDNADNA.ipynb and click on it. You will be directed to a new tab with the exercise. All instructions and background information about the activity will be given in the Jupyter Notebook.
  • Important: If you want to re-connect: Check that you don’t have any open screens with running conda environments by typing screen -r

 
The instructions to the activity can be viewed on github.