Add new tool: PyDrugLogics

[szlaura]

PyDrugLogics is a Python package designed for constructing, optimizing Boolean Models and performs in-silico perturbations of the models.

Features:

• Construct Boolean model from .sif file
• Load Boolean model from .bnet file
• Optimize Boolean model
• Generate perturbed models
• Evaluate drug synergies

Tools already part of colomoto-docker and used in the project:

• MPBN (Most Permissive Boolean Networks)
• PyBoolNet

The GitHub repository for PyDrugLogics avaiable here.

All the neccessary adding tool steps have been completed.

Kindly requesting a review from @pauleve . Thank you!

[pauleve]

Thank you very much for this contribution! This looks really great.
I won't be able to process it until end of next week, I'll keep you in touch as soon as possible.

[pauleve]

Hello,

Thanks again for your pull request, this is a great contribution to the docker image.

I have a few requests to ease the integration:

1. the notebook takes a bit of time to execute. The only concern is with the validation phase of our workflow: each time we upgrade the Docker image, github runs all the notebooks to check that nothing gets broken. Ideally, a notebook should be completed within 1-2min maximum. In your notebook, I see two main points:
   a. the train and predict seem to be executed twice (once with both train+predict, and then separately). Would it be sufficient to show only one way? and mention the other way in text.
   b. the sampling seems to take some time. I would suggest tagging the last cell with "skip_test". It will allow to display the result on the website, but avoid running it during the integration tests

2. there are multiple companion files, notably the "example_models" folder. Is it a necessary input for your notebook? if it is an output, please remove it. Otherwise, it would be better to not put it in the Docker image and instead downloading it from the notebook. You could store the directory as a zip file in your github or on a zenodo record, and use wget or python function to download it and extract it when necessary (see https://colomoto.github.io/colomoto-docker/tutorials/scBoolSeq/scBoolSeq%20-%20scRNA-Seq%20binarization.html for an example)

3. could you please remove the .idea/ folder? and check if all the added files in the tutorials/PyDrugLogics folder are used as input for the notebook.

4. re-run from scratch the full notebook so the numbering of cells and their output is complete

Let me know once you updated the pull request, and do not hesitate to ask for clarification or help!

[szlaura]

Hi @pauleve!

Thank you very much for your helpful comments and suggestions! I finished the necessary changes:

• removed the .idea folder
• removed the repetitive parts of the tutorial
• so that the "example_models" input files are no longer necessary, I removed them too
• tagged the sampling function with "skip_test"
• re-ran the jupyter notebook tutorial

Let me know if you have any more comments or questions!
Have a nice day!

[pauleve]

Great, thank you very much!
It will be shipped in the 2025-01-01 image. Until then, you can use the :next image (which should be ready within 1h), see #155 .