Master Thesis: Accurate quantification of spatial-temporal patterns from high resolution microscopy
Organization: Institut Curie
Many important processes in biology follow spatial or spatial-temporal patterns. Whether a gene is expressed or not might depend on its spatial location at a particular time, and spatial clustering of proteins has a key role in gene expression, cell signaling, etc. These are inherently multi-scale processes: a protein that is clustered in the microscale might not be clustered in the nanoscale, or might be clustered in both scales but differently. With advanced high-resolution microscopy (Nobel prize in 2014), it is possible to study these processes with the required spatial and temporal resolution. Yet, quantitative analysis tools have lagged behind[2,3] in accurately processing the large quantities of noisy data (‘big data’), though several recent attempts have been made (e.g., Stone et al, Nature Communications 2015; Rubin-Delanchy et al, Nature Methods 2015, Baumgart et al Nature Methods 2016). We propose a project that, depending on the candidate’s skills and interests, can combine both programming (preferably Matlab, but open to other languages like Java or Python), and quantitative method development. The candidate will have the possibility to perform experiments to validate the tools developed. The starting point of the project will be the fast, scalable, efficient implementation of existing key spatial/spatial-temporal correlation & cluster analysis methods[3,4,7]. The tools developed will immediately be used by several researchers and groups to study various systems including nuclear organization and epigenetics and also in cell biology in general. We expect the results of the project to be published in high quality journals and open-source software to be released publicly.
The project will be conducted at the LOCCO group, UMR 168, Institut Curie. We are a highly inter-disciplinary group, with a high level of collaboration between physicists, biologists and computer scientists and have a history in publishing software works in high-impact journals such as Nature Methods[8,9].
Skills required: basic mathematics and programming. Recommended skills: signal & image processing, probability & statistics, algorithms, quantitative biophysics
Contact: Arun Shivanandan(firstname.lastname@example.org), Antoine Coulon(email@example.com), Maxime Dahan(firstname.lastname@example.org)
References and suggested reading
1. Betzig, E. et al. Imaging Intracellular Fluorescent Proteins at Nanometer Resolution. Science 313, 1642–1645 (2006).
2. Shivanandan, A., Deschout, H., Scarselli, M. & Radenovic, A. Challenges in quantitative single molecule localization microscopy. FEBS Lett. 588, 3595–3602 (2014).
3. Shivanandan, A., Unnikrishnan, J. & Radenovic, A. On characterizing protein spatial clusters with correlation approaches. Sci. Rep. 6, 31164 (2016).
4. Stone, M. B. & Veatch, S. L. Steady-state cross-correlations for live two-colour super-resolution localization data sets. Nat. Commun. 6, 7347 (2015).
5. Rubin-Delanchy, P. et al. Bayesian cluster identification in single-molecule localization microscopy data. Nat Meth 12, 1072–1076 (2015).
6. Baumgart, F. et al. Varying label density allows artifact-free analysis of membrane-protein nanoclusters. Nat Meth advance online publication, (2016).
7. Coulon, A. & Larson, D. R. Chapter Seven-Fluctuation Analysis: Dissecting Transcriptional Kinetics with Signal Theory. Methods Enzymol. 572, 159–191 (2016).
8. El Beheiry, M. & Dahan, M. ViSP: representing single-particle localizations in three dimensions. Nat. Methods 10, 689–690 (2013).
9. El Beheiry, M., Dahan, M. & Masson, J.-B. InferenceMAP: mapping of single-molecule dynamics with Bayesian inference. Nat. Methods 12, 594–595 (2015).