xpcs_autoencoder

Autoencoder-based automatic analysis of XPCS correlation maps from protein dynamics

The package provides the source code used in the following scientific publication:

[1] Automated matching of two-time X-ray photon correlation maps from protein dynamics with Cahn-Hilliard type simulations using autoencoder networks
S. Timmermann, V. Starostin, A. Girelli, A. Ragulskaya, H. Rahmann, M. Reiser, N. Begam, L. Randolph, M. Sprung, F. Westermeier, F. Zhang, F. Schreiber, and C. Gutt. (2022). J. Appl. Cryst. 55, 751-757 https://doi.org/10.1107/S1600576722004435.

Installation

Requirements

Python 3.7 or above and CUDA 10.2 or above are required. It is also recommended installing PyTorch with torchvision by following the instructions from the official website: https://pytorch.org/get-started/locally/ before installing the xpcs_autoencoder package.

Other python dependencies can be installed automatically during the next step:

• numpy
• scipy
• tqdm
• matplotlib
• h5py
• opencv-python

Install the package

To install the repository locally, execute the following command in the terminal to clone the repository and install it via pip:

git clone [email protected]:schreiber-lab/xpcs_autoencoder.git && cd xpcs_autoencoder && pip install .

or if pip is not available:

git clone [email protected]:schreiber-lab/xpcs_autoencoder.git && cd xpcs_autoencoder && python setup.py install