Protein-Forger

This GitHub repository contains all the dependencies (in yml file) & the necessary files for Proteinforger. You just need to create a conda environment using the yml file.

Create Conda Environment

1. Go to directory where attached openmm.yml is downloaded.

2. Open the terminal & type the following command:

   conda env create -f openmm.yml

3. It would automatically make a conda environment naming openmm

4. Verify the env creation by following command:

   conda env list

Build and Run a Docker Image

If you prefer to run the app in a docker image instead of a conda environment. Then build the Docker image by following these steps:

1. Open a terminal and navigate to the directory containing the Dockerfile.

2. Build the Docker image using the following command:

   docker build -t proteinforger:latest .

3. Once the image is built, you can run the Docker container using the following command:

   docker run -p 8501:8501 proteinforger:latest

This will start the Streamlit app, and you can access it by navigating to http://localhost:8501 in your web browser.

Stopping the Docker Container

To stop the Docker container running the app, follow these steps:

1. List the running Docker containers to find the container ID:

   docker ps

2. Stop the Docker container using the container ID:

   docker stop <container_id>

Replace <container_id> with the actual container ID from the previous command. >

User Manual for Proteinforger

1. Clone the repository & then go to the downloaded directory.

2. Open the terminal in the directory and write the following command:

   conda activate openmm

3. Then run python script through streamlit using following command:

   streamlit run proteinforger.py

   • OR simply run bash script in the terminal:

   ./proteinforger.sh

4. This will start the Streamlit app, and you can access it by navigating to http://localhost:8501 in your web browser.

5. Select the necessary options & do not forget to specify the Project Title.

6. By Default CPU will be used for minimization but if you have multiple proteins, GPU/CUDA option can be selected.

7. Hit the Minimize button.

Note

• Default Force Field will be CHARMM36
• Default Platform will be CPU
• Default # of Minimization steps would be 100. (for small proteins)
• Refer to the excel file ERRAT.xlsx regarding suggested Number of Minimization Steps, for reference.

Author's note: For further assistance email me at muneebgojra@gmail.com

This app was originally built at BioCode.ltd