ZMPY3D_PT

Update: ZMPY3D PyTorch implementation is available (August 25, 2024).

• 3D Zernike moments with PyTorch:
• Shape similarity with PyTorch:
• Structure superposition with PyTorch:

ZMPY3D: accelerating protein structure volume analysis through vectorized 3D Zernike Moments and Python-based GPU Integration

For CPU support only, please refer to the repository:

ZMPY3D supports NumPy (https://github.com/tawssie/ZMPY3D)

For GPU support with TensorFlow, CuPy and PyTorch, please refer to the other three repositories:

ZMPY3D_TF supports Tensorflow (https://github.com/tawssie/ZMPY3D_TF)

ZMPY3D_CP supports CuPy (https://github.com/tawssie/ZMPY3D_CP)

ZMPY3D_PT supports PyTorch (https://github.com/tawssie/ZMPY3D_PT)

Here presents a Python-based software package, ZMPY3D, to accelerate the moments computation by vectorizing the mathematical formulae, enabling their computation in graphical processing units (GPUs). The package offers popular GPU-supported libraries such as CuPy and TensorFlow along with NumPy implementations, aiming to improve computational efficiency, adaptability, and flexibility in future algorithmic development.

Installation

Prerequisites:

• ZMPY3D : Python >=3.9.16, NumPy >=1.23.5
• ZMPY3D_CP: Python >=3.9.16, NumPy, CuPy >=12.2.0
• ZMPY3D_TF: Python >=3.9.16, NumPy >=1.23.5, Tensorflow >=2.12.0, Tensorflow-Probability >=0.20.1
• ZMPY3D_PT: Python >=3.9.16, NumPy >=1.23.5, PyTorch >= 2.3.1

1. Open the terminal
2. Using pip to install the package through PyPI
3. Run pip install ZMPY3D_PT for the installation

Usage

• 3D Zernike moments with Tensorflow:
• Shape similarity with CuPy:
• Structure superposition with NumPy:
• Runtime evaluation:

Performances

A voxel cube with dimensions of 100x100x100 was applied to perform 10,000 3D Zernike moment calculations, using 2 different maximum orders 20 and 40. Execution times for different hardware configurations using TensorFlow, CuPy, and NumPy libraries:

NumPy

TensorFlow

CuPy

Note: m = minutes, s = seconds.

Cache data for order 40

Due to GitHub's file size limitations, follow these steps to download the cache data for order 40 (1.3G) in the ZMPY3D_PT package:

1. Locate Package Folder

• Open your terminal and execute the following command to find the folder of the ZMPY3D_PT package:
• python -c "import ZMPY3D_PT; print(ZMPY3D_PT.__file__)"
• Note the path, which ends with /User/path/ptyhon/site-packages/ZMPY3D_PT/__init__.py.

2. Navigate to Cache Data Folder

• Go to the cache_data folder at the same level as __init__.py file, i.e., /User/path/ptyhon/site-packages/ZMPY3D_PT/cache_data.

3. Download the Cache File:

• Download the 1.3 GB max order 40 .pkl file to the cache_data folder from the link below. https://drive.google.com/uc?id=1RR1rF_5YJqaxNC5AK0Ie_8MswGb0Tttw

Contributing

Feel free to submit pull requests for improvements or bug fixes.

Citation

Lai, J. S., Burley, S. K., & Duarte, J. M. (2024). ZMPY3D: Accelerating protein structure volume analysis through vectorized 3D Zernike moments and Python-based GPU integration. (Bioinformatics Advances, vbae111, https://doi.org/10.1093/bioadv/vbae111)

License

This project is licensed under the GNU General Public License v3.0. You can view the full license here.