Super resolution means to look at things that are impossible to see. The position of the fluorophores is determined using advanced guessing and although well educated, these guesses are always more or less wrong. When super resolution data is used for down-stream analysis it can greatly improve the result to know the magnitude of these errors, i. e. to be certain about the uncertainty. This is especially true when you follow single fluorophores over time and also need to accommodate the fact the particles move in and out of focus. In a recent publication we show that pointwise localization precision can be accurately estimated directly from imaging data using the Bayesian posterior density constrained by simple microscope properties. The figure shows how the method can be used to improve detection of binding events when these are inferred from diffusion measurements of single fluorophores.
Read the complete article in Nature Communications
The software uncertainSPT contains two basic software tools to extract and use localization uncertainty for single particle tracking and localization microscopy.
- 1) EMCCDfit, localization algorithms to estimate particle positions and localization uncertainty from images aquired by an EMCCD camera.
- 1b) EMCCD_dark_count_calibration.m, a calibration routine that extract EMCCD parameters from very low light images (<<1 photons/pixel).
- 2) EMhmm, a variational EM algorithm that performs maximum likelihood inference in a diffusive hidden Markov model, where both motion blur and localization uncertainty are included in the model, through an extension of the Berglund model for single-state diffusion.
The code runs on matlab, with some inner loops implemented in C/C++. Binaries for 64 bit linux and max OS are included.
Get uncertianSPT from GitHub