Membrane pores: a computer simulation of interacting pores analyzed by g1(tau) and g2(tau) correlation functions.

Ion channels in a cell membrane were modeled by a computer simulation of fluctuating pores distributed in a spatial array, a cellular automata. The sum of the currents through such a set of pores models a noise analysis experiment. These currents were analyzed by using the optical correlation functions gn(tau) = (fn(t)fn(t + tau))/(f2(t))n, where f(t) are the current deviations around the mean current and () denotes the time average. These functions can be easily used to determine if the noise is Gaussian. If the noise is not Gaussian, they provide additional information not already contained in the power spectrum. When the pores do not interact with each other, the noise is Gaussian and the power spectrum a Lorentzian. When the pores interact in a strongly cooperative way the noise was still Gaussian and the power spectrum still a Lorentzian, but the usual analysis applied to such a case would over-estimate the single channel conductance. If the kinetics of the pore opening and closing vary on the time scale of the experiment then the relationship g2(tau) = 1 + 2[g1(tau)]2 is no longer satisfied.