Abstract
It is known that retinal ganglion cell (RGC) produces reliable and precise spike pattern for fluctuating current input.However, stochastic RGC model with precise ion channel properties requires significant calculation time for simulation.On the other hand, the method to speed up simulation of stochastic Hodgkin-Huxley (HH) model was recently proposed.The method approximates the stochastic behavior of individual ion channels which has been originally described by Markov model as massive noise term in the ion conductance.However, this method is only tested on the HH model and the effectiveness to other models has not yet been clarified.In this study, we applied this method to RGC model.The RGC model includes seven kinds of ion channels (Na, K, A, h, Ca(T), Ca(L), KCa).We derived noise term for all ion channels.As a result, we succeeded in simulation speed up of simulation 26 times than that of the previous model.We also confirmed that the model reproduces a similar property of spike timing as the previous model.We evaluated the spike timing reproducibility and found that the reproducibility increases as the standard deviation of the noise increases.
