KLT Picker: particle picking using data-driven optimal templates.
Current version: 1.2.8
Project's homepage: https://github.com/ShkolniskyLab/kltpicker
Date: 10/2020
This is the Python version containing the complete source code of the KLT Picker. A MATLAB version is available at https://github.com/ShkolniskyLab/kltpicker_matlab.
Please cite the following paper when using this package: Eldar A, Landa B, Shkolnisky Y. KLT picker: Particle picking using data-driven optimal templates. J Struct Biol. 2020 May;210(2) 107473.
The package has been tested on Ubuntu 16.04 and Windows 10. It should probably work on other versions of Windows and Linux, but has not been tested on them yet. Similarly for macOS.-
Python 3.6.0+ is required.
-
The package makes use of the pyfftw package, which in turn uses the FFTW library. Before installing KLTPicker make sure you have the FFTW library installed on your system: http://www.fftw.org/fftw3_doc/Installation-and-Customization.html#Installation-and-Customization
-
For optional GPU support, the package requires:
- NVIDIA CUDA GPU with the Compute Capability 3.0 or larger
- CUDA Toolkit: v9.0 / v9.2 / v10.0 / v10.1 / v10.2 / v11.0
$ pip install kltpicker
In order to enable the GPU support (provided that your system satisfies the above requirements):
$ pip install kltpicker[gpu]
$ pip install kltpicker-x.x.x.tar.gz
You can also install the development version of KLTPicker from a cloned Git repository:
$ git clone https://github.com/ShkolniskyLab/kltpicker.git
$ cd kltpicker
$ pip install .
$ pip uninstall kltpicker
$ pip install -U kltpicker