A tool for high throughput data processing for live cell
imaging.
0. Introduction
This is the project homepage for the Time Lapse
Analyser (TLA), a free, cross-platform, open source
software for high throughput time-lapse data processing for live cell
imaging. Currently we provide 21 different use cases, ranging from
basic cell counting
and image processing
to wound healing assay
analysis and
cell tracking applications for
differential interference contrast (DIC), phase contrast (PC),
fluorescent imageing (FI) and brightfield microscopy images.
TLA is a graphical tool which enables easy access to high-throughput
live cell imaging for every user, regardless of the individual
expertise. Beginners can easily process stacks of time-lapse data by
loading an apropriate image processing setup and time-lapse evaluation
setup. Advanced users can modify existing setups, or create new setups
which suit their need and special experimental conditions.
The online documentation
provides a quick tour for beginners. Configuration files for predefined
program setups are introduced in the
setup section. The program and documentation are available in
the download
section. The source code is distributed under a Creative Commons
Attribution-Noncommercial 3.0 License and available in section download.
This software was designed to enable fast and easy access for new
users. The basic user mode requires only a minimum
amount of user interaction.
Running a time lapse analysis
The upper graphic shows the basic user mode after
opening the TLA. Only five steps are needed to run a predefined
time-lapse analysis:
Load avi file / set of avis
Load setup file
Choose saving directory
Define file ending
Start processing
Load avi file / set of avis
The first step is to load the set of time lapse files. This is done via
the Open avis / set of avis menu entry or the Open
Avi button. The avi files have to be either uncompressed or
compressed with one of the following formats (Windows only): Indeo 3,
Indeo 5, Cinepak, MSVC, RLE. The loaded files are visible in the drop-down
list on the upper panel. Also the current active file can be changed there by
selecting an item. During batch processing, the current processed file
is shown in the drop down box as the current item as well as printed in
the command window.
Load setup file
The next step is to load a predefined setup file. A setup file
configures the image preprocessing operations and the time lapse
analysis process. Some basic setups are described in section Setup, e.g. a
multi-target-tracking on a set of cells. The advanced user
mode (see Software
documentation) provides a GUI to define additional setup
settings.
Choose saving directory
The directory for file saving is set via the Browse
button.
Define file ending
A unique file ending for the time lapse results can be set in the text
field below the Browse button. All results are
automatically saved under the avi file name extended by the user
defined file ending. E.g. a list of avi files File1.avi, ..., File5.avi
and the result ending _trackingresult will be saved
after processing as File1_trackingresult.mat, ...,
File5_trackingresult.mat.
Start processing
Before processing is started a valid saving directory and result file
ending should be chosen. The default ending is "_trackingresult", the
default saving directory is a folder "tlaresults" automatically created
in the avi folder. Press "Start time lapse" and confirm the pixel
sizes, magnification and recording time in the displayed dialog. After
processing the TLA result viewer will open which provides some useful
tools to display the tracking results, a video tool to superimpose the
tracks into the avi file and some export functions.
Basic example
This section describes a step-by-step example on how to run a basic
time lapse analysis setup configuration. This example setup is used for
multi-target cell tracking in DIC (differential interference contrast)
images. The mandatory cell confluence and image quality is shown by
some example images which can be downloaded together with the setup
file. Download the avi dic_panc1_vid1.avi
as well as the sample setup setup_dic_tracking1.zip
to follow this example.
Load avi file
Start the program and load the example1.avi file.
The upper graphic shows
the basic user mode after loading a video
file.
Load setup file
Unzip the setup_dic_tracking1.zip file and load the
setup_dic_tracking1.mat file
Start processing
Before processing is started a valid saving directory and result file
ending should be chosen. The default ending is _trackingresult, the
default saving directory is a folder tlaresults automatically created
in the avi folder. Press Start time lapse and confirm the pixel sizes,
magnification and recording time in the following dialog. After the
processing the result view will open in which the tracking results can
be saved.
The advanced user mode gives access to various functions for image
enhancing and video configuration. In this user mode a setup editor can
be used to write own setup configurations.
Beneath basic use cases are given which illustrate the power of the
setup editor. Each setup bundle includes at least sample images and a
readme.txt file with detailed setup informations (Image resolution,
recording technique, tested cell type, cell size, etc.).
Cell counting
setup_cellcounting_dic1.zip
Cell counting in DIC (differential interference contrast) images
working with coarse cell region and cell detail segmentation.
setup_cellcounting_dic2.zip
Cell counting in in DIC (differential interference contrast) images
working with deconvolution.
setup_imagestack1.zip
Computes the mean frame from a loaded video file. E.g. to identify
regions in the video frames, without any movement.
setup_imagestack2.zip
Computes the corase area of cells in each frame and adds them together.
Can also be used to identify regions with no cell movement and the
general activity of cells.
Video processing
setup_mov1.zip
Adjusts contrast in the video and samples only every second frame. Also
the frames are resized to 50% of their original size.
setup_mov2.zip
Highlights a certain area of the video and dimishes the rest. Records
with pseudo-colors (jet), samples only every second frame and resizes
the video to 40% of its original size
Single Image processing
setup_img1.zip
Highlights a region of intrest (ROI) in an image. The ROI is defined
manually in the TLA previously to the analysis.
setup_img2.zip
Contrast enhancement for DIC and PC images
setup_proliferation2.zip
Counts the number of proliferations in DIC (Differential Interference
Contrast) images.
Scratch wound assay analysis
setup_woundhealing1.zip
Wound healing assay analysis. Single cells that leave the cluster are
ignored. All holes in the cell area are closed. Gives a smooth cell
border.
setup_woundhealing2.zip
Wound healing assay analysis. Single cells will also be regarded down
to a size of 1500 squared pixels. Holes in the cell area are closed
only if they are smaller than 3000 pixels.
Tube assay analysis
setup_tubes1.zip
Setup for tube formation assay analysis. DIC and PC possible.
setup_tubes2.zip
Setup for tube formation assay analysis. This setup shows the
application of the Region of Interest (ROI) operation. Example shows
results with and without using a predefined ROI.
4. Installation
The software is written in MatLab and the files are compiled
as a stand-alone Graphical User Interface (GUI)
program. This software requires either a MatLab version (>=v.
7.2) or the MATLAB Compiler Runtime (MCR).
Install MCR If you do not have a Matlab version on your system. An MCRInstaller for each platform is included in the program's zip package.
Start TimeLapseAnalyser by
Windows: Double click on tla.exe.
Unix/Linux/MacOS X: Call ./run_tla.sh
<your MCR installation root directory> from
the command line.
Running the source code from within Matlab
If you are using the TLA-source code from within a Matlab environment, make
sure to include all provided folders and subfolders into your Matlab path.
This is most easily done using the genpath() function in
combination with the addpath() function, e.g. addpath(genpath). If you
want to add the paths permanently, use the function savepath() after you added the directories.
If you cannot open your avi video files with TimeLapseAnalyser, it may
be due to an unsupported video codec (on Unix systems, avis must be
uncompressed. On Windows systems supported codecs are: Cinepak, Indeo3,
Indeo5, MSVC). Also, avi files with special characters (like an '.' or '#' ) may not be readable by TLA.
Under Windows, you can use virtualdub (http://www.virtualdub.org/) to
decompress avi files.
Unfortunately, virtualdub is not working with Wine under Unix, but you
can try out Avidemux (http://avidemux.org).
If
you experience problems with the TLA image processing results, check
whether your Avis are compressed (if you are working on a windows
machine). Strongly compressed avis may circumvent the image processing
to work properly. If you have the possiblity to work with uncompressed
material, make use of it. If only compressed material is available, the
setup may have to be adjusted in the advanced user mode.
Acknowledgements
Example Image and video file courtesy of:
Götz v. Wiechert*, Denis Krndija*, Thomas Seufferlein*+ (*Clinic
of Internal Medicine I, Medical Centre Ulm University,
Albert-Einstein-Allee 23, D-89081 Ulm, Germany; +Department of
Internal Medicine I, Martin-Luther-University, Halle-Wittenberg,
Germany).
Benjamin Künemuth,
Brigitte Lankat-Buttgereit, Harald Schmidt, Malte Buchholz (Department
of Gastroenterology and Endocrinology, University Hospital of Marburg,
Germany).
Dr. Sascha Meyer dos Santos (Klinikum der Johann Wolfgang
Goethe Universitaet Institut für Klinische Pharmakologie).
Kristian Mølhave, Cristian Gradinaru (DTU Nanotech - Dept.
of Micro and Nanotechnology, Technical University of Denmark).
