ESPCI web site
Information theoretic analysis of stochasticity in metabolism
Fluctuations in gene expression and growth rate of single cells can have important consequences for cellular function and fitness. In ref [1], fluctuations in the instantaneous growth rate of single cells of Escherichia coli and in the expression of metabolic enzymes have been measured using time-lapse microscopy.
In order to characterize the degree of dependence between these variables beyond correlation functions, we infer from this data information theoretic quantities such as transfer entropy and information flow, the significance of which will be illustrated on some simple analytic examples. Our results highlights the role of the extrinsic component of the noise, which makes the dynamics non-bipartite and leads to previously unexplored regimes for information dynamics. We hope that our method could contribute to general efforts to understand stochastic fluctuations in biological systems.
[1] Stochasticity of metabolism and growth at the single-cell level, D. J. Kiviet et al., Nature, 514, 376 (2014).