MDM2-PFEP-Paper

Instructions for reproducing results from MDM2 alchemical side-chain mutation study.

Installation instructions

General installation

This paper utilizes the BioSimSpace residue-of-interest (ROI) functionallity for creating perturbable protein molecules. This functionality is available in all BioSimSpace versions including and after 2024.2. To install a general BioSimSpace installation:

conda create -n openbiosim -c conda-forge -c openbiosim biosimspace
conda activate openbiosim

Exact paper version

The paper itself uses 2024.2 version of BioSimSpace for the perturbed molecule generation. If you wish to use this specific version, you have several options:

1. Running provided docker image (recommended)
2. Installing older version via conda

Running provided docker image

1. Install Docker into your machine
2. Pull the provided docker image: docker pull akalpokas/biosimspace_2024.2:latest (repository: https://hub.docker.com/r/akalpokas/biosimspace_2024.2)
3. Interactively run the pulled image: docker run -it akalpokas/biosimspace_2024.2:latest

You will be placed in the /home/ubuntu/ directory and in the openbiosim conda environment.

5. And finally run python and import BioSimSpace: import BioSimSpace as BSS

Running provided Jupyter notebooks

If you wish to interact with the Jupyter notebooks obtained from the this GitHub repository, mount the local GitHub repository as a volume in the docker container:

docker run -p 10000:8888 -it -v /absolute/path/to/git/directory/:/home/ubuntu/mdm2 akalpokas/biosimspace_2024.2:latest

Note

You can choose to change the port numbers on your machine if you have other services running on port 10000 for example.

Now once inside the container, change the directory to the mounted volume mdm2 and launch the jupyter lab server:

jupyter-lab --no-browser --port=8888 --ip=0.0.0.0 --NotebookApp.token='' --NotebookApp.password=''

Now you should be able to access the Jupyter lab server outside the docker container on your host machine: http://127.0.0.1:10000/lab or http://localhost:10000/lab

Installing older version via conda

Warning

There is no gurantee of this method working on every machine, especially in the future when some BioSimSpace dependencies might become unavailable due to version deprecation.

1. Install conda/miniconda/miniforge on your machine. For example miniforge is available here: https://github.com/conda-forge/miniforge
2. Install the specific BioSimSpace version: conda create -n openbiosim -c conda-forge -c openbiosim "biosimspace==2024.2.0"
3. Access BioSimSpace by activating the conda enviroment: conda activate openbiosim
4. And run python and import BioSimSpace: import BioSimSpace as BSS

---

Input files

Repository Structure

The input files in this repository is split into 3 different parent folders:

1. 01_prepare_inputs - documents how the raw protein and ligand input files were transformed into input simulation files.
2. 02_run_prepared_inputs - documents how the prepared input files were taken, assembled, and ran via GROMACS.
3. 03_analyse_free_energies - documents how the post-processed simulation data was used to generate error statistics and plots.