Application of the maximum entropy method to absorption kinetic rate processes Original Research Article

The maximum entropy method, originally developed for astronomical image restoration, has already been successfully applied to a variety of biophysical problems. Through numerical inverse Laplace transformation, the method determines the lifetime distribution function with the largest informational entropy. Starting from a flat distribution, it results in the consistent selection of a single distribution from the numerous possible ones that correctly fit the data. In this paper, we discuss the application of the method to kinetic processes that have both rise and decay components, and test the algorithm with different signal to noise ratio generated data. It is proved that the mass conservation constraint can be taken into account by reducing the search to a lower dimensional subspace. The effect of noise on the width of lifetime distribution is studied and it is shown that an inherent entropy connected to the underlying kinetics can be separated from the noise generated entropy. The possibility of the application of the method to the photocycle kinetics of bacteriorhodopsin is also shown.