Maximum entropy analysis of data simulations and practical aspects of time-resolved fluorescence measurements in the study of molecular interactions
Abstract
Time-resolved fluorescence spectroscopy and microscopy is increasingly used to probe molecular interactions and dynamics events in vitro and in vivo. We point out some pitfalls to avoid in the data acquisition procedure using time correlated single photon counting. A good accuracy in fluorescence decay measurements is not only linked to the counts in the peak channel but also to the statistics at the end of the curve. A too narrow time interval between successive excitation leads to an overlap of decays, and that should be taken into account in data analysis. The counting rate in the peak channel of the excitation profile should be close to the one of the fluorescence decay.
Many distributions of lifetimes can fit an incomplete or noisy data set to satisfying precisions corresponding to close values of χ2. The maximum entropy principle is appropriate to distinguish among these in a consistent way. It is also shown that encoding a prior knowledge about the system study dramatically improves the quality of the recovered distribution, particularly in case of a set of discrete components. Based on simulated noisy quantify maximum entropy method (MEM) data analysis, we propose a simple strategy for estimating the quality of information and inferences we can draw from experimental results.- Publication:
-
Journal of Molecular Structure
- Pub Date:
- December 2014
- DOI:
- 10.1016/j.molstruc.2013.12.079
- Bibcode:
- 2014JMoSt1077...77H