Abstract:In this paper, we present criteria for non-exponential ergodicity of continuous-time Markov chains on a countable state space. These criteria can be verified by examining the ratio of transition rates over certain paths. We applied this path method to explore the non-exponential convergence of microscopic biochemical interacting systems. Using reaction network descriptions, we identified special architectures of biochemical systems for non-exponential ergodicity. In essence, we found that reactions forming a cycle in the reaction network can induce non-exponential ergodicity when they significantly dominate other reactions across infinitely many regions of the state space. Interestingly, the special architectures allowed us to construct many detailed balanced and complex balanced biochemical systems that are non-exponentially ergodic. Some of these models are low-dimensional bimolecular systems with few reactions. Thus this work suggests the possibility of discovering or synthesizing stochastic systems arising in biochemistry that possess either detailed balancing or complex balancing and slowly converge to their stationary distribution.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: how to determine whether a continuous - time Markov chain has non - exponential ergodicity on a countable state space, and apply it to chemical reaction systems to explore the non - exponential convergence behavior of microscopic biochemical interaction systems. Specifically, the author proposes a path - based method to verify the non - exponential ergodicity of continuous - time Markov chains. By examining the ratio of transition rates on certain paths, these conditions can be verified. In addition, the author applies this method to chemical reaction networks, identifies special architectures that lead to non - exponential ergodicity, and discovers some biochemical systems of detailed balance and complex balance, which slowly converge to their steady - state distributions. The following are the key points in the paper: 1. **Problem Background**: - A continuous - time Markov chain (CTMC) is usually exponentially ergodic on a finite irreducible state space. - However, in a countably infinite state space, there are cases where exponential ergodicity is not guaranteed, especially in stochastically modeled biochemical systems. 2. **Proposed Method**: - The author proposes new sufficient conditions to prove the non - exponential ergodicity of continuous - time Markov chains on a countable state space. - These conditions indicate that if there are a series of paths in the state space such that the Markov chain will stay for a long time once it enters these paths, then the chain is non - exponential ergodic. 3. **Application Areas**: - Apply these conditions to chemical reaction networks to identify the types of biochemical systems with non - exponential ergodicity. - In particular, when there are reactions forming cycles in the reaction network, the dominance of these reactions in the network will lead to non - exponential ergodicity. 4. **Theoretical Basis**: - Use path probabilities and moment - generating functions to quantify the behavior of Markov chains on specific paths. - Propose the concept of a strong tier - 1 cycle to describe the structural features in the reaction network that lead to non - exponential ergodicity. 5. **Conclusions and Prospects**: - Through the above methods, the author has found many different non - exponential ergodic stochastic biochemical systems, which are either in detailed balance or in complex balance. - The study also discusses the relationship between these results and the spectral gap and congestion ratio, which are usually used to prove exponential ergodicity. In summary, this paper aims to provide a new method to analyze and identify biochemical systems with non - exponential ergodicity, which is of great significance for understanding the dynamic behavior of complex biochemical processes.

A path method for non-exponential ergodicity of Markov chains and its application for chemical reaction systems

A new path method for exponential ergodicity of Markov processes on $\mathbb Z^d$, with applications to stochastic reaction networks

Non-explosivity of endotactic stochastic reaction systems

A Stein's Method Approach to the Linear Noise Approximation for Stationary Distributions of Chemical Reaction Networks

Markovian Dynamics on Complex Reaction Networks

Explicit exponential convergence to equilibrium for nonlinear reaction-diffusion systems with detailed balance condition

Stationary Equations for Non-Markovian Biochemical Systems

A Scalable Computational Framework for Establishing Long-Term Behavior of Stochastic Reaction Networks

Exploring Dissipative Sources of Non-Markovian Biochemical Reaction Systems.

A Stochastic Framework for the Design of Transient and Steady State Behavior of Biochemical Reaction Networks

Boundary-induced slow mixing for Markov chains and its application to stochastic reaction networks

Classification and threshold dynamics of stochastic reaction networks

Complex balanced distributions for chemical reaction networks

Product-Form Poisson-Like Distributions and Complex Balanced Reaction Systems

A Moment-Convergence Method for Stochastic Analysis of Biochemical Reaction Networks

Convergence to equilibrium of renormalised solutions to nonlinear chemical reaction-diffusion systems

Stochastic Chemical Reaction Networks with Discontinuous Limits and AIMD processes

Trend to equilibrium for reaction-diffusion systems arising from complex balanced chemical reaction networks

Long-time asymptotics of stochastic reaction systems

Noise and critical phenomena in biochemical signaling cycles at small molecule numbers

Stochastic Theory of Nonequilibrium Steady States. Part II: Applications in Chemical Biophysics