Many cellular functions depended on the mechanical interactions of cells with their substrate. Traction force microscopy measures the forces that cells exert on their substrate. Additionally the mechanical properties of whole cell colonies, can be analyzed using a Finite Elements approach (Monolayer Stress Microscopy). These properties are foe example the distribution of stresses across the colony surface, as well as the forces transmitted across cell-cell contacts. Here both methods are implemented in a clickpoints addon, making it easy to quickly analyzed whole datasets.
In traction force microscopy cells are seeded on a linearly elastic substrate and allowed to adhere to its surface. The substrate, e.g. Polyacrylamide Gel, is filled with micrometer sized fluorescent beads. When the cell exerts forces on its substrate it causes deformations. These deformations can be tracked by imaging the beads. First an image of the beads is taken. Then the cells are detached from their substrates, e.g. by trypsinization. Now another image of the beads is taken. The two images of the beads are used to calculate a deformation field with Particle Image Velocimetry. The traction forces i.e. forces applied from the cells to the substrate surface are then calculated using Fourrier Transform Traction Force Microscopy (FTTC). Earlier implementations of FTTC relied on assuming infinite substrate thickness. Here a correction for finite substrate thickness is included. Forces that are exerted from a cell to its substrate must be balanced by the cell internally or at contact points to other cells. The internal stress state of a cell patch is calculate by Monolayer Stress Microscopy. In brief the cell sheet is modeled as a 2D surfaces. The traction forces calculated from FTTC are applied to this cell sheet. Internal stresses are then recovered by using a standard 2D Finite Elements approach. The user can mark cell borders,along which line stresses i.e. the force transmitted per line segment, are calculated. Additionally stress measures across the whole colony area, such as the average shear and normal stress are also calculated.
Standard Fourrier Transform Traction Force Microcopy:
Traction fields, moments, and strain energy that cells exert on their surroundings
James P. Butler, Iva Marija Tolić-Norrelykke, Ben Fabry, and Jeffrey J. Fredberg
Am J Physiol Cell Physiol 282: C595–C605, (2002)
Fourrier Transform Traction Force Microcopy with finite substrate thickness:
Physical forces during collective cell migration
Xavier Trepat, Michael R. Wasserman, Thomas E. Angelini, Emil Millet, David A. Weitz,
James P. Butler and Jeffrey J. Fredberg
Nature Physics volume 5, pages 426–430 (2009)
Monolayer Stress Microcopy:
Monolayer Stress Microscopy: Limitations, Artifacts, and Accuracy of Recovered Intercellular Stresses
Dhananjay T. Tambe, Ugo Croutelle, Xavier Trepat, Chan Young Park, Jae Hun Kim, Emil Millet,
James P. Butler, Jeffrey J. Fredberg
PLOS ONE 8(2): e55172 (2013)
It is recommended to use this package with anaconda. If you are on Windows you need the Microsoft Visual C++ build tools. Download and install them from [here] (https://visualstudio.microsoft.com/de/thank-you-downloading-visual-studio/?sku=BuildTools&rel=16). Additionally you need to install clickpoints. See here for instructions.
Next install the traction force microscopy package. I recommend to make an editable installation with pip. Simply download or clone the repository. Unizp the files, then open a terminal and navigate to the folder tracktion_force_mircoscopy. Now perform a local installation with the command
pip install -e .
You can also install this package directly from github. First you need to install git. For windows you can use
conda install git
then install this package by
pip install git+https://github.com/fabrylab/tracktion_force_microscopy.git
This will automaticaly add the an addon to clickpoints if you have clickpoints 1.9.0 or higher.
For general instructions on how to use clickpoints go here. First open a clickpoints database, e.g by right clicking on an image and "open with" clickpoints. Addons can be activated by pressing on the central right most button. A window listing all available addons will open. Select "TFM_addon" and press activate. You should get a massage that the addon has been succesfully activated. Note that you can also activate other useful addons here. One example is the "Measure Tool", used to measure distances, or for example the size of your beads.
After you have activated the Addon a new button appears on the right. Press this button to open the addon window.
If you have opened a new database you first need to tell the addon where it can find images and where it should put the output files. Press the "select images" button. This will open a new window:
In this window (left side), you can tell the addon which folders it should go through in order to search for images that are taken before and after cell removal. On the right, you can enter a regular expression pattern to identify specific images in the folders you have selected, for example flurescent images of the beads, or bright field images of the cells. For example (see the screen snapshot above), images of the beads after cell removal need to contain a number with 1-4 digists, followed by the word "after". The images for the beads after cell removal also need to contain a number with 1-4 digists, followed by the word "before", and so on. You also need to tell the addon where in the filename it can find the frame number of the image. For example, "^(d{1,4})" means the frame number is at the very beginning of the file name, and it contains between 1-4 digits.
Here are some other common regular expressions you migth enter in these fields: "after"
| search pattern | meaning |
|---|---|
| after | all files with "after" in the filename |
| * | all files |
| for the frames | |
| ^(\d{1-4}) | up to 4 numbers at beginning of the filename |
| (\d{1-4}) | up to 4 consecutive numbers anywhere in the filename |
| (\d{1-4})$ | up to 4 numbers at end of the filename |
Note the the file needs to have an extension for common image formats (.png, .jpeg, .tif and so on)
You can also set a folder for all output files and specify a name for the database. The database will be saved to the output folder. Once you have set all options press the "collect images" button. Now the addon searches for images, sorts them to the database and saves the database. The programm will print all images that it found, what frame and what type it identified for them to the console. Also if there are more or less then 3 images per frame you will see a warning in the console.
You can also setup your database form the python interpreter if you prefer to do so:
Either open python in the terminal or use the interpreter for example in PyCharm. Import the function to setup a data base:
from pyTFM.database_functions import setup_database_for_tfm
set the folder that your images are located in. It could look like this:
folder = r"C:\Users\Andy\Desktop\exmaple_analysis\KO_shift_parts"
call the setup_database_for_tfm function with default parameters on this folder. You have to provide the folder and a name for your data base. The name has to end with ".cdb".
setup_database_for_tfm(folder,"database.cdb")
You can also pass your own keys to sort images as parameters to the function. For each parameter you need to provide a regular expression. The regular expression to identfy a frame needs to contain the actual frame number as a group marked with "()". Don't include the file ending. ".png" or ".tif" and so on is added automatically.
# using custom search keys for the data base setup:
key1="\d{1,4}after" # finds any file name with 1 to 4 numbers followed by "after"
key2="\d{1,4}before" # finds any file name with 1 to 4 numbers followed by "before"
key3="\d{1,4}membrane" # finds any file name with 1 to 4 numbers followed by "membrane"
frame_key= "(\b\d{1,4})" # finds the first 4 leading numbers of the filename
# build the data base:
setup_database_for_tfm(folder,"database.cdb",key1=key1,key2=key2,key3=key3,frame_key=frame_key)
Slight movements of the microscopes field of view while removing cells will reuslt in non-zero drift in the deformation field. You can use this addon to correct for any shift between the images of the beads before and after the removal of cells. Once you have set up the databse in the "select file" menue, just press the correct drift button below it:
This will correct the drift for all frames using image registration. All three images of a frame will be cut to a common field of view. This will permanetly change your image files. Note that this function can not correct for roations.
In the top right you can tick which part of analysis you want to run. The raw output from these analysis, such as the deformation field, is stored as an array in the same folder. That way it can be accessed later by other analysis steps. Deformation and traction fields will also be plotted and added to your database in a new layer. Depending on the analysis mode "FEM analysis" or the "contractillity analysis" will also produce an image and added it to the clickpoints database. You can view the images by simply changing layers in your current frame in clickpoints. During the analysis several measures (area of cells, contractile energy and so on) are calculated. They are all stored in an text file called out.txt. If there is already an out.txt file from aprevious analysis, the new file will be named out0.txt, out1.txt and so on. The field "apply to" allows you to run the analysis on just the current frame or all frames at once. To start your analysis press the start button on the top left.
You can choose between two different analysis modes: "cell layer" and "colony". "Cell layer" assumes that the whole field of view is covered in cells. Finite Elements analysis is performed with nodes at the edge of the field of view fixed, so that they can't move orthogonal to the image edge. You are supposed to mark two different areas ("cell type1" and "cell type2"). On these areas you average stresses and average contractile energy are calculated. "contractillity_measures" will add a map of the contractile energy to the database. A fixed area close to the image edges ( default 10% of the image axis length) is excluded for the calculation of all measures. When you select "cell layer" a new button "fill cell area" appears. This button helps you by trying to fill all areas encircled with the mask for cell type1 on all frames. Alternatively clickpoints also provides a tool to fill areas individually. has to masks for two cell types.
In the "colony" mode you can analyze an isolated cell colony. The finite elements analysis will not fix any node (provided you don't mark an area at the image edge). Instead the equilibrium of the system is guaranteed by correcting nodal loads for unbalanced forces and torque. You are supposed to mark the colony edges and internal cell cell borders within the colony. Additionally you can circle an area around the colony to calculate contractillity and contractile energy. "FEM analysis" in this mode will produce stress measures along the cell borders and on the whole colony area. It will also add an image of the cell border stresses to the database. Note that the FEM analysis is only accurate if you have a high resolution in deformation and traction field. This can be achieved by choosing the "piv overlap" close to the "piv window size". Eventually you should set this difference as low as 5 pixels or less (1 µm at 0.2 µm pixelsize) . Unfortunately this will increase your calculation time.
All measured quantities are saved to an output file named "out.txt". This file is only generated when you perform the analysis on all frames of the database. The parameters used for the analysis are written as header. All measured quantities follow with the frame, name of the quantity, it's value, a unit and an additional warning message, each in one column. You should reconsider the results if you see a warning.
This package provides some functions to read the output file, perform statistical test comparing two output files and plotting these results. Check out the analysis in output_data_analysis for a detailed example on how to compare the analysis results results for two cell types. You can find a minimal example of a completed analysis in the example_analysis folder. It compares the forces exerted by cell colonies of wild type and plectin knock out cells. Note that I change the path entry in the database so that they work immediately when you download this folder.
Default parameters and plotting options can be found in the [file parameters_and_string] (/pyTFM/parameters_and_strings.py). Go to this file for more details. The parameters are mostly stored in dictionaries. You can freely change entries in these dictionaries, if you installed your package as editable. Note that currently reinstalling the package will overwrite changes you make there.
Make sure you draw closed circles around the cell colony and the area selected for contractillity. The program tries to fill the encircled areas, but will fail if there is a gap. If you get a warning, that your mask is tool small, this could be the reason. Also any cell border in the cell colony mask that is not connected at both ends with another border is removed.
If there is a problem with the finite elements analysis, try increasing the resolution of deformation and traction field.
The images in your are linked to the database with an absolute path. If you move an image or change the name of the folder the images are in, they will no longer be displayed. You can of cause edit the path entry of the database.