Stochastic dynamics of two-compartment models with regulatory mechanisms for hematopoiesis

Ren-Yi Wang,Marek Kimmel,Guodong Pang

2024-05-06

Abstract:We present an asymptotic analysis of a stochastic two-compartmental cell proliferation system with regulatory mechanisms. We model the system as a state-dependent birth and death process. Proliferation of hematopoietic stem cells (HSCs) is regulated by population density of HSC-derived clones and differentiation of HSC is regulated by population density of HSCs. By scaling up the initial population, we show the density of dynamics converges in distribution to the solution of a system of ordinary differential equations (ODEs). The system of ODEs has a unique non-trivial equilibrium that is globally stable. Furthermore, we show the scaled fluctuation of the population converges in law to a linear diffusion with time-dependent coefficients. With initial data being Gaussian, the limit is a Gauss-Markov process, and it behaves like the FCLT limit under equilibrium with constant coefficients at large times. This is proved by establishing exponential convergence in the 2-Wasserstein metric for the associated Gaussian measures in a $\mathcal{L}_2$ Hilbert space. We apply our results to analyze and compare two regulatory mechanisms in the hematopoietic system. Simulations are conducted to verify our large-scale and long-time approximation of the dynamics. We demonstrate some regulatory mechanisms are efficient (converge to steady state rapidly) but not effective (have large fluctuation around the steady state).

Populations and Evolution

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to study the efficiency and effectiveness of two different regulatory mechanisms in the hematopoietic system by analyzing the stochastic dynamics of a two - compartment cell proliferation system with regulatory mechanisms. Specifically, the author focuses on how the self - renewal and differentiation processes of hematopoietic stem cells (HSCs) are regulated by cell density, and the dynamic behaviors of these regulatory mechanisms on large - scale and long - time scales.

### Background and Objectives of the Paper
- **Background**: Hematopoiesis refers to the process of blood cell generation. Hematopoietic stem cells can self - renew, differentiate, or die. When a hematopoietic stem cell differentiates, it will become a multipotent progenitor cell and further differentiate until it produces mature cells, forming a clone that will eventually become extinct. Regulatory mechanisms are required to maintain the homeostasis of the hematopoietic system (i.e., the stability of the cell generation system).
- **Objective**: This paper aims to establish a framework for analyzing the efficiency and effectiveness of the hematopoietic stem cell proliferation system in a two - compartment model with regulatory feedback. The author pays special attention to the large - scale approximation of system dynamics (Functional Law of Large Numbers, FLLN) and the fluctuations around the equilibrium state (Functional Central Limit Theorem, FCLT) when the initial population size is large.

### Main Contributions
1. **Model Construction**:
- The author models the hematopoietic system as a state - dependent birth - death process, in which the self - renewal rate \(\alpha\) of hematopoietic stem cells depends on the density \(\bar{N}_1\) of the derived clone, and the differentiation rate \(\gamma\) of hematopoietic stem cells depends on the density \(\bar{N}_0\) of hematopoietic stem cells.
- By scaling the initial population size \(r\), the author shows that the dynamics of cell density converges in distribution to the solution of a system of ordinary differential equations (ODEs) with a unique non - trivial globally stable equilibrium point.

2. **Theoretical Analysis**:
- **Functional Law of Large Numbers (FLLN)**: It is proved that the dynamics of cell density weakly converges to a deterministic function \(\bar{N}\) as \(r\rightarrow\infty\), and this function satisfies a system of ordinary differential equations with a globally stable non - trivial equilibrium point.
- **Functional Central Limit Theorem (FCLT)**: Consider the difference between the cell density dynamics and the FLLN limit, and scale this difference by \(\sqrt{r}\) to obtain a linear diffusion process \(\hat{N}\), whose time - dependent coefficients become constant when the FLLN limit starts from the equilibrium state.

3. **Application and Verification**:
- The author applies this framework to analyze two specific regulatory mechanisms and verifies the dynamic approximations of these mechanisms on large - scale and long - time scales through simulation.
- The results show that although some regulatory mechanisms can converge to the steady state quickly (efficient), the fluctuations around the steady state are large (inefficient).

### Key Formulas
- **System of Ordinary Differential Equations (FLLN)**:
\[
\begin{cases}
\bar{N}'_0(t)=\alpha(\bar{N}_1(t))\bar{N}_0(t)-\gamma(\bar{N}_0(t))\bar{N}_0(t)\\
\bar{N}'_1(t)=\gamma(\bar{N}_0(t))\bar{N}_0(t)-\delta\bar{N}_1(t)
\end{cases}
\]
where \(\bar{N}_0(0)=\bar{n}_0\), \(\bar{N}_1(0)=\bar{n}_1\).

- **Linear Diffusion Process (FCLT)**:
\[
d\hat{N}(t)=A(t)\hat{N}(t)dt+\sigma(t)dB(t)
\]
where
\[
A(t)=\nabla F(\bar{N}(t)),\quad\sigma(t)=\begin{pmatrix}
\sqrt{\alpha(\bar{N}_1(t))\bar{N}_0(t)}
\end{pmatrix}