ABSTRACT
AIM: To study quantitatively the memory existing in ion channels. METHODS: Stochastic processes were used to model 2 categories of memory (short-term and long-term) by persisting in the standpoint of two-state, instead of multiple states, but with different transition mechanism. RESULTS: A two-state Markov process with constant transition intensities well fitted the short-term memory and a two-state Markov process within a kind of random environment well fitted the long-term memory. Statistical procedures for parameter estimation were proposed and demonstrated with 2 real examples on the channels of PC12 cells. CONCLUSION: The memory in ion channels can be quantitatively modelled as stochastic process with 2 states.
Subject(s)
Markov Chains , Memory, Short-Term , Memory , Adrenal Gland Neoplasms/pathology , Humans , Ion Channels , Nerve Growth Factors/pharmacology , Pheochromocytoma/pathology , Tumor Cells, Cultured/drug effectsABSTRACT
AIM: To explore the existence of memory in Ca(2+)-dependent K+ channels of cultured aortic smooth muscle cells (ASMC) and voltage-dependent K+ channels of clonal pheochromocytoma (PC12) cells. METHODS: Calculating the sample auto-correlation functions based on the digitized signals or the 0-1 series corresponding to closing and opening of the channels after routine evaluation, rather than the sequence of sojourn times. RESULTS: The sample auto-correlations showed a decreasing trend with elapse of time, stable to repeated observations under the same conditions and sensitive to treatments. CONCLUSION: The attribute of memory exists in some single channels as an intrinsic feature of them, independent of any extrinsic assumptions on missing observations due to limited time resolution.
