cryo3d is a matlab-based software that allows for fast 3D protein reconstruction based on cryogenic particle images. The software was developped at Image Processing and Analysis Group, Yale School of Medicine, Dept. of Diagnostic Radiology under supervision and with participation of Prof. Hemant Tagare.
/doc : contains project description-related documents
/script : main *.m scripts of the pipeline and main workspace (the Matlab should be run from that folder)
/src : *.m functions, arranged by the different pipeline step, e.g., preprocessing, best_match, etc; mrc folder contains function to read-write *.mrc files
/test : temporal functions that are under development, subject to delete for the final release version
To download the source code of the pipeline, run the following commands:
path $ git clone https://github.com/vicrucann/cryo3d
To get the submodules (cacharr and rshell-mat):
path $ cd cryo3d
path/cryo3d $ git submodule update --init --recursive
When planning to edit (and then push back to the github) any of the submodules, checkout to the master branch for each of the submodules:
path $ cd cryo3d/src/rshell-mat
path/cryo3d/src/rshell-mat $ git checkout master
path/cryo3d/src/rshell-mat $ cd ..
path/cryo3d/src $ cd cacharr
path/cryo3d/src/cacharr $ git checkout master
User must provide configuration file as an input for the \script\cryo3d.m function (you can find an example of such file for both Windows and Linux in script folder.
1. (*.star, *.mrcs) -> fit_ctfs.m -> fit-ctfs.mat
2. (fit-ctfs.mat, *.mrcs) -> proprocess_images.m -> pre-imgs.mrcs
3. (*.mrc) -> init_volume.m -> ini-vol.mrc
4. (theta) -> coordinate_axes.m -> coord-ax.mat
All of function for fast best match method were written by Nicha Dvornek. The improvements like main framework organization, distributor package rshell-mat and caching class cacharr were written by Victoria Rudakova. For inquires on additional packages: vicrucann(at)gmail(dot)com.