Kimura Mutsumi, Akimoto Takuma
Department of Physics, Tokyo University of Science, Noda, Chiba 278-8510, Japan.
J Chem Phys. 2023 Aug 7;159(5). doi: 10.1063/5.0156073.
Local diffusivity of a protein depends crucially on the conformation, and the conformational fluctuations are often non-Markovian. Here, we investigate the Langevin equation with non-Markovian fluctuating diffusivity, where the fluctuating diffusivity is modeled by a generalized Langevin equation under a double-well potential. We find that non-Markovian fluctuating diffusivity affects the global diffusivity, i.e., the diffusion coefficient obtained by the long-time trajectories when the memory kernel in the generalized Langevin equation is a power-law form. On the other hand, the diffusion coefficient does not change when the memory kernel is exponential. More precisely, the global diffusivity obtained by a trajectory whose length is longer than the longest relaxation time in the memory kernel is not affected by the non-Markovian fluctuating diffusivity. We show that these non-Markovian effects are the consequences of an everlasting effect of the initial condition on the stationary distribution in the generalized Langevin equation under a double-well potential due to long-term memory.
蛋白质的局部扩散率关键取决于其构象,且构象涨落往往是非马尔可夫性的。在此,我们研究具有非马尔可夫涨落扩散率的朗之万方程,其中涨落扩散率由双阱势下的广义朗之万方程建模。我们发现非马尔可夫涨落扩散率会影响全局扩散率,即当广义朗之万方程中的记忆核为幂律形式时,通过长时间轨迹得到的扩散系数。另一方面,当记忆核为指数形式时,扩散系数不变。更确切地说,由长度长于记忆核中最长弛豫时间的轨迹所得到的全局扩散率不受非马尔可夫涨落扩散率的影响。我们表明,这些非马尔可夫效应是由于长期记忆,初始条件对双阱势下广义朗之万方程中的稳态分布产生持久影响的结果。