Calcium (Ca 2+) is an ubiquitous intracellular messenger which regulates cellular processes, such as secretion, contraction, and cell proliferation 1;2.Avariety of cell types respond to hormonal stimuli with periodic oscillations of the intracellular free
of entropy rates is common to the theory of stochastic processes as well as to the information theory where the entropy rates are used to characterize information production by information sources 3 . Alternatively, the time series fytg can be considered as a projection of a trajectory of a dynamical system, evolving in some measurable state space. A. N. Kolmogorov, who in
troduced the theoretical concept of classi cation of dynamical systems by information rates 4, was inspired by the information theory and generalized the notion of the entropy of an information source 4;5;8. The Kolmogorov-Sinai entropyKSE 5;6;7;8 is a topological invariant, suitable for classi cation of dynamical systems or their states, and is related to the sum of the system's positive Lyapunov exponentsLE according to the theorem of Pesin 9 . Thus, the concept of entropy rates is common to theories based on philosophically opposite assumptionsrandomness vs. determinism and is ideally applicable for characterization of complex biological processes, where possible deterministic rules are always accompanied by random in uences. In this paper we demonstrate an application of so-called coarse-grained entropy rates for the characterization and segmentation of high-frequency oscillations of intracellular Ca2+ Ca2+ i in clonal cellsinsulin secreting hamster cells induced by a hormonal agonistarginin-vasopressin, AVP and synergistically activated by increasing doses of an additional agonisttolbutamide.
2 Information-Theoretic Functionals and Entropy Rates Consider n discrete random variables X1;:::; Xn with sets of values1;:::;n, respectively. The probability distribution for an individual Xi is pxi= PrfXi= xi g, xi 2i . We denote the probability distribution function by pxi, rather than pXixi, for convenience. Analogously, the joint distribution for the n variables X1;:::; Xn is px1;:::; xn= PrfX1;:::; Xn=x1;:::; xng,x1;:::; xn 21:::n . The marginal redundancyX1;:::; Xn,1; Xn, in the case of two variables also known as mutual information IX1; X2, quanti es the average amount of information about the variable Xn, contained in the n, 1 variables X1;:::; Xn,1, and is de ned asX1;:::; Xn,1; Xn= X::: X px;:::; x log px1;:::; xn:1 1 n px1;:::; xn,1pxn x1 21 xn 2n
热门文档