Simulations of rotational diffusion ("tumbling") of fluorophores

This repo contains GPU-accelerated code for modeling tumbling of fluorophores. This code supports our recent paper, "From cameras to confocal to cytometry: measuring tumbling rates is a general way to reveal protein binding." Read the publication here: https://andrewgyork.github.io/tumbling_is_general

Citation:

If you use this code in your research, please cite the following paper:

Julia R. Lazzari-Dean, Austin E.Y.T. Lefebvre, Rebecca Frank Hayward, Lachlan Whitehead, Maria Ingaramo, Andrew G. York. From cameras to confocal to cytometry: measuring tumbling rates is a general way to reveal protein binding. (2023)

Instructions:

To replicate the data, figures, and animations from the paper, follow these steps:

1. Follow the setup instructions below.
2. Navigate to the respective "get_figures" directory for the figure you would like to replicate.

• For data replication: Run the simulation_"figure_name".py file to generate the data for the figure.
  • This data will be saved in the "data" subdirectory within the figure directory.
• For plotting replication: Run the plot_"figure_name".py file to generate the figure.
  • Note: This automatically loads the latest data from the "data" directory.
• For animation replication: Run the animate_"figure_name".py file to generate the animation.
  • The gif animations and the individual frames will be saved in the "images" subdirectory within the figure directory, with subdirectories based on 3 orthogonal projection-like views and a skewed 3d view.

Setup:

Packages:

• Note: it's recommended but usually not required to use a clean virtual environment to install the packages below to avoid conflicts with other packages on your system.

  • Learn more about virtual environments here: https://docs.python.org/3/tutorial/venv.html

• CPU-based (slower for large simulations):

  • option 1 (recommended): pip install -r requirements.txt
  • option 2: pip install numpy pandas matplotlib

• GPU-based (slower for small simulations)

  • Windows and Linux with NVIDIA GPUs only
  1. Install CUDA if you haven't already:
     • Windows: https://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/index.html
     • Linux: https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html
  2. Install required packages:
     • option 1 (recommended): pip install -r requirements.txt cupy-cuda12x
     • option 2: pip install numpy pandas matplotlib cupy-cuda12x
       • Note: cupy-cuda12x is the latest version of cupy that supports CUDA 12.0, but you should install the latest version of cupy that supports your specific CUDA version.

System Requirements:

Everything in this repo has only been tested on the following:

• python 3.10
  • Mac
  • Windows
    • CPU
    • GPU
  • (It probably also works on Linux for both CPU and GPU)