This plugin allows to use cryoCARE -trains a denoising U-Net for tomographic reconstruction according to the Noise2Noise training paradigm- tomography methods into Scipion framework.
The plugin can be installed in user (stable) or developer (latest, may be unstable) mode:
1. User (stable) version::
scipion3 installp -p scipion-em-cryocare
2. Developer (latest, may be unstable) version::
- Clone the source code repository:
git clone https://github.com/scipion-em/scipion-em-cryocare.git
- Install:
scipion3 installp -p local/path/to/scipion-em-cryocare --devel
The integrated protocols are:
- Load a previously trained model.
- Generate the training data.
- Training: uses two data-independent reconstructed tomograms to train a 3D cryoCARE network.
4. Predict: generates the final restored tomogram by applying the cryoCARE trained network to both even/odd tomograms followed by per-pixel averaging.
The installation can be checked out running some tests. To list all of them, execute:
scipion3 tests --grep cryocare
To run all of them, execute:
scipion3 tests --grep cryocare --run
The test generates a cryoCARE workflow that can be used as a guide about how to use cryoCARE. The even/odd tomograms required to use cryoCARE can be generated inside Scipion with:
- Plugin scipion-em-motioncorr: protocol "align tilt-series movies".
- Plugin scipion-em-xmipptomo: protocol "tilt-series flexalign".
- Cryo-CARE: Content-Aware Image Restoration for Cryo-Transmission Electron Microscopy Data. Tim-Oliver Buchholz et al., 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019).
If you experiment any problem, please contact us here: scipion-users@lists.sourceforge.net or open an issue.
We'll be pleased to help.
Scipion Team