This package includes several tools to study conformational heterogeneity from CryoEM data.
- Zernike3Deep: Semi-classical neural network to analyze continuous heterogeneity with the Zernike3D basis
- HetSIREN: Neural network heterogeneous reconstruction for real space
- ReconSIREN: Neural network for ab initio reconstruction and global angular assignment
- FlexConsensus: Consensus neural network for conformational landscapes
- Annotate space: Interactive inspection of conformational landscapes and conformational states
- MoDART: Real space reconstruction with motion correction derived from deformation fields
Hax needs pip to install its dependencies. The installation on a independent Conda environment is strongly recommended.
We recommend installing the package directly from Pypi using the command:
# Cuda 13 command
pip install hax-em[cuda13]
# Cuda 12 command
pip install hax-em[cuda12]
If you prefer to have a local copy of this repository, you may also clone directly from GitHub and install the package with the following command (assuming that you are already inside the cloned folder):
pip install .
Warning
Supported NVIDIA drivers version: >= 525 (Cuda 12/13 will be installed along the package, so there is no need to have CUDA already installed in your system).
After installing the software, you will be able to launch the command hax_project_manager. This command allows to easily execute all the tools integrated inside Hax package.
A detailed explanation on how to call the programs can be obtained by running:
hax_project_manager -h
To see the help of any program, you may use:
hax_project_manager program -h
Where program is the name of any method listed by the first help command (e.g. hetsiren)
- Herreros, D., Lederman, R.R., Krieger, J.M. et al. Estimating conformational landscapes from Cryo-EM particles by 3D Zernike polynomials. Nat Commun 14, 154 (2023).
- Herreros, D., Kiska, J., Ramirez-Aportela, E. et al. ZART: A Novel Multiresolution Reconstruction Algorithm with Motion-blur Correction for Single Particle Analysis. Journal of Molecular Biology 435, 168088 (2023).
- Herreros, D., Mata, C.P., Noddings, C. et al. Real-space heterogeneous reconstruction, refinement, and disentanglement of CryoEM conformational states with HetSIREN. Nat Commun 16, 3751 (2025).
- Herreros, D., Perez Mata, C., Sanchez Sorzano, C.O. et al. Merging conformational landscapes in a single consensus space with FlexConsensus algorithm". Nat Methods (2023).
