Retourner en haut

Team : SIBARITA

Research Projects

Quantitative analysis of super-resolution data

Project Leader(s): Florian Levet

We have developed and published a suite of software dedicated to the analysis, quantification and visualization of SMLM data. We also developed a novel technique for quantifying the spine morphology in images acquired in STED microscopy. All these solutions are distributed under different license agreement (free, open-source, collaborative, commercial).

PALM-Tracer & WaveTracer are software platform allowing the localization, tracking and visualization of multi-color SMLM data. It includes a patented method which uses a combination of wavelets and Gaussian fitting for the 3D localization of single molecules, and simulated annealing algorithm for their tracking (Izeddin et al., Optics Express 2012, Kechkar et al., Plos One, 2013). Distributed under collaborative agreement, tens of laboratories and imaging centers worldwide are using it in routine. It was recently used to decipher a new important activation mechanism of b-arrestins, a regulator protein of G-proteins coupled receptors (Eichel et al., Nature, 2019). WaveTracer is a commercial plugin licensed to Molecular Devices since 2013. It allows the analysis and reconstruction of SMLM data online during the acquisition. It plays a central role in the development of the HCS-SMLM screening platform (Beghin et al., Nature Methods, 2017)

SR-Tesseler is a stand-alone software platform allowing the efficient segmentation of SMLM data using Voronoï polygons centered on the localized molecules (Levet et al., Nature Methods, 2015). We use the Voronoi polygons as a way to add relevant descriptors to the localizations; and in particular to compute a local normalized density that can be used to automatically segment and analyze datasets displaying very different localization densities. We also used it to determine the organization of pre- and post-synaptic proteins (Chamma et al., Neurophotonics, 2016) and to automatically quantify clusters acquired with our HCS-SMLM platform (Beghin et., Nature Methods, 2017). Packaged with an adapted visualization and first released as a Windows installer, its code-source is available since 2019 under a GPLv3 license (commercial licensing is possible). Since its publication in Sept. 2015, it was downloaded more than 1,500 times and cited more than 200 times, and has become one of the reference methods in the field.

Coloc-Tesseler is our last standalone software platform (Levet et al., Nature Communication, 2019). It uses overlays of Voronoï polygons computed on each color to compute standard colocalization coefficients, such as Manders, Spearman or object‘s distances, from 2D and 3D multicolor SMLM data. Similarly to SR-Tesseler, the use of Voronoi diagrams makes the technique robust to variation in the datasets densities. Currently available as a Windows Installer, its code-source should be soon available.

SpineJ is a plugin for ImageJ allowing morphological analysis of spine geometry acquired in STED microscopy (Levet et al., Methods, 2020). It uses a combination of wavelet filtering, vector field, Delaunay triangulation and graph theory to facilitate quantitative analysis of the spine neck and head. It was successfully used in several collaborations, for analyzing spines (Letellier et a., PLoS Biology, 2019, Inavalli et al. Nature Methods, 2019, Angibaud et al., Journal of Physics D: Applied Physics, 2020) and astrocytes (Arizono et al., Nature Communications, 2020).

Fundings

None

Back to the Team