Supervisory authorities

CNRS


Presentation of the institute

Search




Home > TEAMS > TEAM SIBARITA > SOFTWARE

SR-Tesseler

A method to segment and quantify localization-based super-resolution microscopy data


Software download




News


2017/05/22:

  • Addition of the barycenter of each object and cluster in the xls files generated by the export of the statistics.


2017/04/19:

  • Addition of the size of the major and minor axis for objects obtained with the DBSCAN method (useful for non-isotropic objects). It is also possible to choose between PCA and the bounding ellipse method to estimate the size of these objects.


2016/11/14:

  • Added the possibility to snap the viewer in vectorial form (svg file). While the export works well for points and lines, a well-known issue of vector files with triangles is the fact that their boundaries is not filled (http://stackoverflow.com/questions/15631426/svg-fill-not-filling-boundary). As a solution, the user can define the stroke width used with triangles to fill their boundaries. This width is thus dependant of the zoom level of the snap.

  • Added the possibility to estimate the size of objects/clusters with a bounding ellipse in addition to the PCA (principal component analysis). Please note that the bounding ellipse estimation is a lot longer than the PCA one.


2016/10/13:

  • Added support for the H-function (linearized and normalized version of the K-Ripley’s function) and DBSCAN. Those two techniques are available in a new tab called "Misc quantification".


2016/05/31:

  • Fixed a minor issue caused by miscommunication between Qt and Windows that may cause unexpected crashes of the software when running while doing nothing.


2016/05/25:

  • Added the possibility to export the features (local density, mean distance and area) computed on the Voronoi polygons. To achieve this task, a new button (with a floppy disk icon) has been added below the histogram of the feature.


2016/04/29:

  • Corrected a name error on the ImageJ plugin that prevented its compilation. Added the .class to the archive too.


2016/04/25:

  • Added a new export for the objects/clusters. In this export, each localization is represented as a pair: the first column is the number of the localization while the second column is the number of the object or of the cluster it is part of (0 if this localization is part of the background).


2016/04/15:

  • Added the possibility to change the location and name of the export files for statistics performed on objects and clusters. Added the possibility to export to clipboard the results tables of objects and clusters statistics too.


2016/01/27:

  • Added the support for localization files generated by Zeiss ZEN software. In general, the loading of the localization file is more general and the SR-Tesseler should manage to identify the separator (comma, tabulation or space). For csv files, the headers are fixed though. For x coordinates it should be called "x" or "Position X", for y coordinates "y" or "Position Y", for intensity "intensity" or "Number Photons", for time "frame" or "First Frame" and for localization error "sigma" or "Precision".


2015/11/19:

  • Added the possibility to change the maximum dark time for the blinking correction (textbox next to the "Fixed max dark time" checkbox). This value will determine the number of bins of the histograms (fitted with the different curves) that will determine the dark time used for the correction of the localizations. If the checkbox is checked, this user-defined dark time will be used for reconnection instead of the one computed by the histograms.
  • Added the possibility to export the different histograms used for the determination of the maximum dark time. Three files ("tons.txt", "toffs.txt" and "blinks.txt") will be created in the directory of the localization file.


2015/11/04:

  • Fixed a bug preventing the possibility to change the fixed neighborhood size for the blinking correction.
  • Addition of the possibility to add the standard deviation (sigma) of the PSF emitter when fitted with a Gaussian model for each localization (see the updated manual for the file format -.txt and .csv - specifications). These sigmas are needed for the "photon size" and "photon/background size" blinking correction.
  • Addition of a text box to define the ratio between the intensity of the localization and the corresponding photon number.
  • Change of the structure used to display the blinking parameters in order to have scrollbars.


2015/11/03:

  • Fixed a bug leading to the crash of the software upon creation of the objects.


2015/10/28:

  • Added the support for csv files as input files for the localizations. The file needs a header with at least "x", "y", "intensity" and "frame" columns.
  • Added the possibility to save and load the actual position as well as zoom level of the viewer. Useful if you want to get back at exactly the same position as the one used for a previous analysis.
  • Added the possibility to perform a snapshot of the current viewer. Snapshots can be saved as uncompressed jpg, png or tif files.
  • Added a 64 bits installer for SR-Tesseler. As of right now, the 32 bits installer is abandoned. If you are using the 32 bits SR-Tesseler version and want to use the 64 bits one, make sure to uninstall SR-Tesseler before running the installation of the 64 bits version.

Abstract


Localization-based super-resolution techniques open the door to unprecedented analysis of molecular organization. This task often involves complex image processing adapted to the specific topology and quality of the image to be analyzed.

SR-Tesseler is an a freely available segmentation software using Voronoï tessellation constructed from the coordinates of localized molecules. It allows precise, robust and automatic quantification of protein organization at different scales, from the cellular level down to clusters of a few fluorescent markers.

SR-Tesseler is insensitive to cell shape, molecular organization, background and noise, allowing comparing efficiently different biological conditions in a non-biased manner, and perform quantifications on various proteins and cell types.

SR-Tesseler software comes with a very simple and intuitive graphical user interface, providing direct visual feedback of the results and is freely available under GPLv3 license.


(A) Diffraction limited fluorescence image of a cultured neuron expressing GluA1-mEOS2. (B) PALM super-resolution image reconstructed from the localized molecule coordinates. (C) Voronoï diagram constructed from the localized molecule coordinates. Color of the polygons encodes for the local density computed from the Voronoï polygons. (D) Multi-scale segmentation obtained using SR-Tesseler revealing the cellular contour as well as molecular clusters.



Software download

SR-Tesseler is an interactive software platform that allows multiscale segmention and quantification of 2D localization microscopy data. It is developed in C++ and is using a few extern libraries:

SR-Tesseler is developed for a Windows environment and comes with no waranty of any kind. Two installers (32 & 64 bits) are available:

Source code is available on demand, after signing a material transfer agreement.

For any question or request or if you want to be informed of any new version, feel free to send a message to: