Lateral drift correction without fiduciary markers in PALM experiments - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

Lateral drift correction without fiduciary markers in PALM experiments


diffraction limit, practically a few tens of nanometers. Thousands of images are acquired, each of which containing individual image of just a few fluorophores, ideally well separated. This acquisition takes a few seconds or minutes to be acquired, a time during which the sample experimentally drifts. The drift is often a few tens of nanometer, which is on the order of the expected resolution, and thus need to be corrected. Drift correction is generally estimated by randomly adding beads of dyes whose fluorescence lasts all the acquisition. Then, the experimental drift is estimated by following these beads, for example using a fitting method. The major drawbacks are the step to add beads, and the facts that for good drift estimation, there should be a good compromise of the number of beads (3/4) and beads should not bleach before the end of the acquisition. To face these problems, we propose method without fiduciary markers to estimate lateral drift in PALM experiments. The algorithm principle is based on subpixel localization of the center of correlation maps between substacks of images. Many parameters are important for precise drift estimation. The importance of each parameter will be presented using both simulated and experimental actin-lifeact-TdEos2 cytoskeleton PALM data. We will also focus on the effect of different algorithms to preprocess the stack of data.
No file

Dates and versions

hal-00763829 , version 1 (11-12-2012)


  • HAL Id : hal-00763829 , version 1


Laurent Guyon, C. Mignon, Fabrice Senger, Laurent Blanchoin, J. Xie, et al.. Lateral drift correction without fiduciary markers in PALM experiments. Swiss Single Molecule Localization Microscopy, Aug 2012, Lausanne, Switzerland. ⟨hal-00763829⟩
205 View
0 Download


Gmail Mastodon Facebook X LinkedIn More