Quick Start
This page describes in a simplified way the first steps required to start using MicrobeTracker and MicrobeTracker tools. See also more detailed tutorials based on the included datasets. Overall, the tutorials are listed in the order of increasing complexity:
Additionally, the help pages for some tools have extended examples of their usage:
Contents
MicrobeTracker requires MATLAB with Image Processing Toolbox to run, though it can be compiled and will run stand-alone. However, MATLAB will still be needed to view or process the output files and run the tools. Here are the instructions how to install the MATLAB-based version:
Unpack or copy the content of MicrobeTracker package into any folder.
Add this folder to the list of default paths in MATLAB (in MATLAB select File>Set Path>Add folder, then select the MicrobeTracker folder and click Save)
Now you can run MicrobeTracker by typing microbeTracker in MATLAB workspace window and the rest of the functions by typing the function's name.
Optionally, you can add support for images in other formats than series of single-page TIFF files and multi-page TIFF files by installing Bioformats software (the images in this case have to be in separate stack files for each filter block / color). To do so, exit MicrobeTracker, download loci_tools.jar file, and place it to the MicrobeTracker folder.
For program and functions description see help either by clicking Help button in MicrobeTracker (for the program itself) or opening help (htm) file from the MicrobeTracker folder (for the rest of the functions).
Sometimes MicrobeTrackers's compiled functions will not run because of some libraries missing on your operating system. To test, type in MATLAB command window a=intxy2C(0,0,0,0), which should return an empty matrix, not an error. If it returns an error, install the compatible C++ compiler, which can be found on MathWorks website (e.g. for version 2011a; make sure you select your operating system).
Prepare the image files to load. If you are using the program for the first use, select images with very clearly visible and well separated cells. This description will focus on using phase contrast microscopy (called simply phase below) images, see the detailed help for diffuse fluorescence or DIC. The files must be in TIFF format placed in a separate folders for phase and corresponding fluorescence images. If the files have to be in a Make sure that the particular order (e.g. timelapse), nake sure that they are sorted correctly by the operating system. Usually the files have the same base name followed by a number. In this case the numbers have to be x01, x02, ... , x09, x10, x11, ... as opposed to x1, x2, ... , x10, ...
Load the phase images by clicking 'Load phase data' button on the top panel and selecting the folder with the images. If fluorescence images are used as well, load them clicking 'Load signal 1' button.
Load parameter set by clicking 'Load parameters' button from the Parameters panel. Typically select 'alg4.set' from the MicrobeTracker folder. It is based on algorithm 4, designed to work with any rod-shaped cells, any curvature and length. See the detailed help to figure out what algorithm to use for specific cases.
Adjust the most critical parameters by modifying the text
in the 'Parameter' control on the bottom right:
- set areaMin and areaMax so that the area of all good cells falls into the
range between these two values. If not known precisely, set areaMin much smaller
and areaMax much larger, and then adjust them more presizely (the area of a
detected cell can be seen on the 'Information' panel on the bottom left once the
cell is selected).
- set cellwidth to the width of the cell in pixels (approximately).
- for filamentous cells increase fsmooth, to measure
variations in cell width reduce wspringconstant, for
curved cells reduce rigidityB, for timelapses with
dividing cells set splitThreshold to ~ 0.4, see the
detailed help for other parameters.
If fluorescence needs to be added, subtract the background by selecting 'All' and clicking 'Subtract bgrnd' button on the 'Background subtraction' panel (bottom center) and select 'Compute signal 1 profile' on the 'Detection & analysis' panel.
Select 'Time lapse' or 'Independent frames' mode on the 'Detection & analysis' panel.
Click on 'This frame' button to see the performance on one frame. Wait for up to a few minutes to process the frame.
If the detection was good, click 'All frames' to process the whole image set. This may take up to several hours, depending on the number and size of the images.
After the program finishes working, save the results by clicking 'Save analysis' button.
This section describes how to use the functions for cell dimensions and intensity statistics. This description is primarily for those who have no MATLAB experience in order to start using the functions. If you are familiar with basic MATLAB concepts, refer directly to the description of the functions.
In order to compare data from two experiments on the same plot you will need cellList from both loaded at the same time. For this rename this variable after loading, for example:
In order to specify additional parameters, put them after cellList. Most of
the other parameters are optional and the order is not important. If the order is
important (when the function cannot distinguish the parameters based on the
variable type), place an empty array [] to omit a parameter. For example, to show
an overlapping histogram (i.e. plotting the overlaps of bars with another color),
type:
meanwidthhist(cellList1,cellList2,'overlap')
To specify the x-axis (an array of lengths of
the cell in which the bin centers will be placed), type:
meanwidthhist(cellList1,cellList2,5:10:100)
The record 5:10:100 means that the centers will be places in positions from 5 to 100 pixels with the step of 10. Usually specify the position of the first bin center at one half of the bin width, so that the bins start from zero.