search

nbcrrolls / txbr-source

Transform based Tracking, Bundle adjustment and Reconstruction (TxBR)
Other
2 stars 0 forks source link

readme

Transform based Tracking, Bundle adjustment and Reconstruction (TxBR)

DOI

Transform based Tracking, Bundle adjustment and Reconstruction (TxBR) is an electron tomography package developed on top of the IMOD utilities. At this time, it does not offer any fiducial tracking options and it loosely follows the ETomo reconstruction scheme.

This source tree was derived from updated versions of TxBR code found here:

https://confluence.crbs.ucsd.edu/display/ncmir/TxBR

Compatibility

Dependencies

Simply put, a lot of software is needed to get this working. Below is a list of the software needed, but it is highly recommended that either the Virtual machine route or the Singularity route be used to use TxBR Both of these approaches are described in the Installation section below.

Quickstart TxBR on the cloud

Click launch button to spin up the latest release of TxBR on the cloud (~10 minute spin up time): (Oregon region)

Launch TxBR AWS CloudFormation link

Click here for detailed instructions on launching TxBR via AWS CloudFormation

Installation

To try TxBR via Vagrant virtual machine out follow these instructions

OR

To build a Singularity image of TxBR do the following, assuming Singularity is installed:

git clone https://github.com/nbcrrolls/txbr-source.git
cd txbr-source
make singularity
dist/txbr-v3.1.2-dev.img --help

OR

Assuming all dependencies have been installed and setup.cfg paths are correct, then the following should work on Centos 7:

git clone https://github.com/nbcrrolls/txbr-source.git
cd txbr-source
. /etc/profile.d/modules.sh
module load mpi/openmpi-x86_64
export LD_LIBRARY_PATH=/lib64:/usr/lib64/mpich/lib:/usr/local/IMOD/lib:/usr/local/IMOD/qtlib
python setup.py build
sudo python setup.py install

Usage

As a starting point for reconstructing volumes, three files need to be provided for each series: preali files, rawtlt files and fid files (cf IMOD). For instance, if we have two series called basenamea and basenameb, a total of six files should be accessible to the TxBR calculations: basenamea.preali, basenamea.fid, basenamea.rawtlt, basenamea.preali, basenamea.fid and basenamea.rawtlt.

It is very important to note that because TxBR jointly process multiple series, the fiducial files of all the disctinct series are related. The marker list should be the same in all the .fid files, even though one fiducial marker might be not visible in a certain series.

To run the TxBR reconstruction, just type in from the command line: 


runtxbr.py -b basename

In the case of mutiple series of the same specimen, a comma separated list of the basenames has to be provided to the script runtxbr.py; for instance in case of a dual tilt series, one should type:


runtxbr.py -b basenamea,basenameb

Alignment of the micrographs, reorientation of the specimen reconstruction, filtering, backprojection and series combination will then follows in an automatic sequence; the final volume will then be displayed in an IMOD windows. It is also possible to run manually each of the reconstruction steps.

A series of options are available for the different TxBR scripts. Type:


runtxbr.py --help

for a list of the options available to the runtxbr script. In particular, to start taking into account of the electron beam curvilinearity, you can set the n2 flag to an integer value higher than 1 (default value for a projective model); in practice, do not set a value larger than 6 for n2.

You can find a test dataset in the examples/data folder of the TxBR directory. Files needed to reconstruct a dual tilt series with TxBR are included in the tar file fhv6.tar.gz.

For more infomration visit the wiki:

https://github.com/nbcrrolls/txbr-source/wiki

Bugs

Please report them here

License

See license in this file: LICENSE.txt

Credits & Acknowledgements