This is a MATLAB package for denoising CTF -affected cryo-EM images based on the following manuscripts: 1) Denoising and Covariance Estimation of Single Particle Cryo-EM Images Tejal Bhamre, Teng Zhang, Amit Singer http://arxiv.org/abs/1602.06632 2) Fast Steerable Principal Component Analysis Zhizhen Zhao, Yoel Shkolnisky, Amit Singer http://arxiv.org/abs/1412.0781 The folder kn_rankest includes code for rank estimation by S. Kritchman and B. Nadler. DEPENDENCIES ---------------- This package should be used in conjunction with the cryo-EM tool box ASPIRE (http://spr.math.princeton.edu/) and will be included in the latest version of ASPIRE. It also requires the NUFFT package (to be included in the latest version of ASPIRE) available at http://www.cims.nyu.edu/cmcl/nufft/nufft.html. INSTRUCTIONS ---------------- 1) Download and install ASPIRE from http://spr.math.princeton.edu/ following the instructions for installation. 2) Add ASPIRE files in your MATLAB path using initpath.m in ASPIRE. 3) If this package is in a separate location than ASPIRE, add the package to your MATLAB path using cwf_paths.m 4) Enjoy the example simulation scripts in cwf_scripts to denoise images. INSTRUCTIONS FOR NUFFT --------------------------- wget http://www.fftw.org/fftw-3.3.4.tar.gz tar -zxf fftw-3.3.4.tar.gz cd fftw-3.3.4 ./configure --prefix=$HOME/local --enable-openmp --enable-shared make make install Then we do the same thing for NFFT: wget https://www-user.tu-chemnitz.de/~potts/nfft/download/nfft-3.2.3.tar.gz tar -zxf nfft-3.2.3.tar.gz cd nfft-3.2.3 ./configure --prefix=$HOME/local --enable-openmp --with-matlab=/usr/local/Matlab/R2013a --with-fftw3=$HOME/local make make install To run everything in MATLAB, we type addpath([getenv('HOME') '/local/share/nfft/matlab/nfft']); addpath([getenv('HOME') '/local/lib']); and check by running simple_test; In case of issues or questions, please email Tejal (tbhamre@princeton.edu) and Jane (jzhao@cims.nyu.edu).