https://drive.google.com/file/d/1MsQyG9i_7IV_GmpuONwPb1bTnmAs1BeG/view?usp=share_link - introduction recording presented via Google Meet’s (in Polish)

First connection and tmux session

Task 1: Using tmux (Terminal Multiplexer)

tmux allows you to create, manage, and navigate between multiple terminal sessions from a single window. Here are the steps to attach, detach, and exit a tmux session without naming the sessions.

Practical Steps

1. Start a New tmux Session

   tmux
   

   • This creates and attaches a new unnamed tmux session.

2. Detach from the tmux Session

   • Press Ctrl+b followed by d.
   • This detaches the session, leaving it running in the background.

3. Reattach to the tmux Session

   tmux attach
   or
   tmux a
   

   • This reattaches to the most recently detached tmux session.

4. Exit the tmux Session

   • Type exit and press Enter.
   • This will close the terminal session. If it's the only or last session, tmux will terminate.

Task 2: Using cat & tail to display a CSV File

The tail command is used to display the end of a file. In the context of CSV files, it can be used to skip the header row.

Practical Steps

1. Create a Sample CSV File

   echo -e "xnat_id,date,project,subject,session,\\nfc9d9ac6-e50c-4d,2024-04-11,MW24a,S2WIO304,,\\nfda17c32-2e88-45,2024-06-23,MW24a,S2WIO158,TP2,\\nfe8f99be-d955-4f,2024-06-23,MW24a,S2WIO870,TP2,\\nff7543be-c3a8-46,2024-06-17,MW24A,S2WIO68,TP2," > sample.csv
   
   cat sample.csv
   

   • This creates a file named sample.csv with a header and two rows of data. “cat” will display its content in Terminal window.

2. Display the CSV File Skipping the Header

   tail -n +2 sample.csv
   
   

   • This command skips the first line (header) and displays the rest of the file.

Task 3: Using gnu parallel and tail to Display Specified Columns of a CSV File

gnu parallel is a shell tool for executing jobs in parallel. You can use it with tail to process and display specific columns from a CSV file.

Practical Steps

1. Display Specified Columns

   • Let's say we want to display only the xnat_id and subject columns.

   tail -n +2 sample.csv | parallel --colsep ',' echo {1} {4}
   
   

   • tail -n +2 data.csv skips the header.
   • parallel --colsep ',' specifies that the columns are separated by commas.
   • echo {1} {4} prints the first and third columns (Name and Occupation).

2. Fake fmriprep analysis

   • Let’s create fake fmriprep execution script

   echo 'echo fmriprep /project \\
            /project/derivatives \\
            participant \\
            --fs-license-file /home/user/freesurfer.txt \\
            --participant-label $2 \\
            --nprocs 6 --mem 10000' \\
   				 > run_fake.sh
   	
   chmod a+x ./run_fake.sh
   

   • chmod a+x makes the script executable.

3. Feed the subjects’ list into fake fmriprep script

   • Let’s create fake fmriprep execution script

   tail -n +2 sample.csv | parallel --colsep ',' ./run_fake.sh {1} {4}
   

   This is how we run mriqc, fmriprep, hcp and many others pipelines. It takes just a few commands to glue it together and run efficiently in parallel on a HCP.