> For the complete documentation index, see [llms.txt](https://docs.virtual-flow.org/tutorials-af/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.virtual-flow.org/tutorials-af/tutorial-2-afvs-on-slurm/the-completed-workflow/extracting-and-viewing-the-results.md).

# Extracting and viewing the results

Docked ligands are stored in archives located in the `/output-files` folder (for example):\
\
/AFVS/output-files/qvina02\_rigid\_receptor1/logs/1/0.tar.gz\\

In order to extract the docked poses and generate a file containing predicted affinities for each compound, one can use the following scripts:

{% file src="/files/8LAxzBLrjS4RAZJ4BMx5" %}

{% file src="/files/A4O9nmtk8FAYHBKlXIX9" %}

The first script to execute is `extract_poses_to_folder.py.` It is meant to extract docking poses located in archives present in the current folder to a folder named `/results.` You can easily modify the path to the archive(s) you want to extract by editing the line 35 in the script:\
\
`Specify the path to your tar.gz file`

`tar_file_path = '0.tar.gz'`\
\
The second script, extract\_scores.py, will generate a file named `scores.csv`, containing the names and binding affinities of all the previously extracted poses, located in the `./results` folder. Please be aware that the script currently only extracts docking scores that have negative values.\
The resulting file can be viewed in the linux terminal using for example `vi` or `nano`.