Coder Social home page Coder Social logo

paulmenzel / motioncorr Goto Github PK

View Code? Open in Web Editor NEW

This project forked from jianglab/motioncorr

0.0 2.0 0.0 13.2 MB

motion correction for movies recorded using direct electron detectors

CMake 0.19% Python 0.35% C++ 97.21% Makefile 0.06% C 0.42% Objective-C 0.19% Cuda 1.58%

motioncorr's Introduction

This Git repositiory of MotioCorr is based on the MotionCorr 2.0 release with the following extensions:
	1. Add a new option -nrw to use running average of neighor frames for correlation analysis to detect relative shifts among frame pairs. It can make the final motion trajectory smoother and presumably more reliable
	2. Add a new option -flp to flip the frames along Y-axis (i.e. vertical direction) before correlation analysis
	3. Read any image format (DM4, mrcs, and essentially all formats used in cryo-EM fields by using EMAN2 IO functions)
	4. Automatically convert to native endian of current computer regardless of starting endianness (i.e. from big-endian used by DM4 images to little-endian used by Intel CPUs)
	5. Compile on all major computer systems (Linux, Mac, and Windows) using the cross-platform CMake build system

**********************************************************************
The contents below are original contents included in MotionCorr 2.0
**********************************************************************

Motion Correction for Dose-Fractionation Stack
Version 1.0, April 11, 2013


Requirement:
    GPU hardware version: >2.0
    GPU memory: >1.5Gb
    CUDA 5.0 or higher
    g++ 4.6 or lower


Compile:
    Run "make" in src folder
 
Features:
GPU enabled. The computations are accelerated by GPU. Processing a stack with
20~30frames only takes ~20s, and usually less than 1 minute for more frames,
hence is suit for on-the-fly processing.

    
Run program:
    1) Run gpuinfo in bin folder to get a list of availabe GPU. If device 0 isn't the GPU you want to use, you need to specify its ID. See the following example.
    2) Run dosefgpu_driftcorr without fowllowing any parameters to get a help.
    3) To process your K2 image, see the following example:
    
       Example 1:   dosefgpu_driftcorr yourstack.mrc
       Example 2:   dosefgpu_driftcorr yourstack.mrc -bin 2 -bft 150 -scc 1
       
    4) You can run dosef_logviewer in the folder you run the program to check the result. 
       
    The program has a set of default output filename, so you don't need to set filename. But you need put all stack file into the same folder you run the program. Or you can specify the output filename in paramereters.
    The binning uses Fourier cropping, so no worry to the binning effects.
    
Note:
    1) A bfactor 150 or 200pix^2 is good for most cryoEM image with 2x binned super-resolution image. For unbined image, a larger bfactor is needed.
    2) Suggest to use more than 20 frames.   
    3) -fod defines the minmum distance of frame comparision, which counts from the middle frame. Suggest to set -fod to TotalNumFrame/4
    
The Motion Correction program was written by Xueming Li at Yifan Cheng
Laboratory at UCSF. For more details, source code and future updates, please visit the
lab website at http://cryoem.ucsf.edu , or contact the author directly.

The Motion Correction program are licensed under the terms of the GNU Public
License version 3 (GPLv3).


***************************************
Update:
07.01.2013  Fix a bug of "align to middle"

08.26.2013  Fix bugs and add new functions:		
				1) support rectangular image, replace -crd with -cdx and -cdy
				2) fix bug of "align to middle", and now can align to any specified frame 
				3) output un-corrected stack
				4) remove -sub option
				5) adjust output function to make the program more stable
				6) the min,max,mean value of output MRC stacks are set to the values of last frame
				
08.31.2013  Add new functions:
            1) Gain and dark normalization
            2) Add non-standard MRC format: mode 5(unsigned 8-bit integer) and mode 6(unsigned 16-bit integer)

***************************************

motioncorr's People

Contributors

wjiang avatar

Watchers

James Cloos avatar Paul Menzel avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.