A method of analyzing nonstationary ionic channel current fluctuations in the presence of an additive measurement noise.

Abstract

A method of estimating the parameters of nonstationary ionic channel current fluctuations (NST-ICF's) in the presence of an additive measurement noise is proposed. The case is considered in which the sample records of NST-ICT's corrupted by the measurement noise are available for estimation, where the experiment can be repeated many times to calculate the statistics of noisy NST-ICF's. The conventional second-order regression model expressed in terms of the mean and variance of noisy NST-ICF's is derived theoretically, assuming that NST-ICF's are binomially distributed. Since the coefficients of the regression model are explicitly related to not only the parameters of NST-ICF's but also the measurement noise component, the parameters of NST-ICF's that are of interest can be estimated without interference from the additive measurement noise by identifying the regression coefficients. Furthermore, the accuracy of the parameter estimates is theoretically evaluated using the error-covariance matrix of the regression coefficients. The validity and effectiveness of the proposed method are demonstrated in a Monte Carlo simulation in which a fundamental kinetic scheme of Na+ channels is treated as a specific example.