Authors: F. Lodesani, M. Bennati, C. Fiegna and M. Tartagni
Affilation: ARCES, Italy
Pages: 563 - 566
Keywords: Monte Carlo, ion channels, algorithm, stochastic processes
The capability of having ligand-gated ion channels embedded in planar biomimetic membranes, together with a proper electronic system for elaborating the signal will give the possibility of creating high-performances sensors, able to detect low concentrations of specific target molecules in fluid mixtures. This paper proposes an algorithm for treating stochastic electrical signals generated by a limited number of channels, in presence of strong background noise. It has been conceived to be computationally suitable for implementation in integrated electronic systems and does not require a-priori assumptions on the number of inserted channels, since it numerically performs a two-parameters estimation: total number of inserted channels and probability of the open-state for the single channel, that is dependent from the molar concentration of the target molecule . Estimation is carried out by a two-steps approach in which signal is analyzed both in the frequency and in the time domain. The algorithm, tested using Monte Carlo simultors, could achieve an estimation of open state probability with a mean percent error less than 1% also for low values of probability (less than 0.1) and in presence of noise having a standard deviation as high as 50% of the single channel current.