Stochastic biophysical modeling of irradiated cells

The paper presents a computational stochastic model of virtual cells irradiation, based on Quasi-Markov Chain Monte Carlo method and using biophysical input. The model is based on a stochastic tree of probabilities for each cell of the entire colony. Biophysics of the cells is described by probabilities and probability distributions provided as the input. The adaptation of nucleation and catastrophe theories, well known in physics, yields sigmoidal relationships for carcinogenic risk as a function of the irradiation. Adaptive response and bystander effect, incorporated into the model, improves its application. The results show that behavior of virtual cells can be successfully modeled, e.g. cancer transformation, creation of mutations, radioadaptation or radiotherapy. The used methodology makes the model universal and practical for simulations of general processes. Potential biophysical curves and relationships are also widely discussed in the paper. However, the presented theoretical model does not describe the real cells and tissues. Also the exposure geometry (e.g., uniform or non-uniform exposure), type of radiation (e.g., X-rays, gamma rays, neutrons, heavy ions, etc.) as well as microdosimetry are not presently addressed. The model is focused mainly on creation of general and maximal wide mathematical description of irradiated hypothetical cells treated as complex physical systems.

💡 Research Summary

The paper introduces a computational framework for simulating the irradiation of a colony of virtual cells using a stochastic tree of probabilities combined with a quasi‑Markov chain Monte Carlo (MCMC) approach. Each cell is represented as a branching process in which possible events—DNA damage, repair, fixation of mutations, cell death, transformation into a cancerous state, and others—are assigned probability distributions supplied as model inputs. By iterating the Monte‑Carlo sampling along the branches, the method generates time‑dependent state trajectories for every cell, effectively capturing the full stochastic dynamics of radiation‑induced biological responses.

A distinctive feature of the work is the adaptation of nucleation theory and catastrophe theory, concepts from statistical physics, to describe the dose‑response relationship for carcinogenic risk. The authors show that, when the probabilities governing mutation fixation and transformation are combined with these theories, the resulting risk curve takes a sigmoidal shape: a relatively flat region at low doses (reflecting adaptive response and repair capacity) followed by a steep increase once a critical dose threshold is crossed. This formulation goes beyond the traditional linear‑no‑threshold (LNT) or simple exponential models by explicitly incorporating a phase‑transition‑like behavior that mirrors experimental observations of non‑linear dose‑response.

Two biologically important phenomena are explicitly embedded in the model: (1) adaptive response, where low‑dose exposure enhances cellular repair mechanisms and reduces susceptibility to subsequent higher doses, and (2) the bystander effect, in which signals from irradiated cells induce DNA damage or death in neighboring non‑irradiated cells. Both are represented by additional probability parameters that modulate the branching process for cells directly hit by radiation and for those that are not, allowing the simulation to reproduce the spatially coupled nature of radiation damage observed in vitro.

The input parameter set includes cell‑cycle‑specific damage probabilities, repair efficiencies, mutation fixation rates, death thresholds, and dose‑dependent damage spectra. While these parameters can be calibrated from experimental measurements for different radiation types (X‑rays, γ‑rays, neutrons, heavy ions, etc.), the present study assumes a generic set of values for a hypothetical cell line, deliberately avoiding the complexities of real tissue architecture, heterogeneous dose distributions, and microdosimetric considerations. Consequently, the model is positioned as a universal mathematical scaffold rather than a detailed tissue‑level predictive tool.

Simulation results demonstrate that the stochastic framework can reproduce a range of phenomena: (i) low‑dose radio‑adaptation manifested as reduced overall cell death, (ii) a rapid rise in cancer transformation probability at doses above the sigmoidal inflection point, (iii) the emergence of mutation clusters and clonal expansion under repeated exposures, and (iv) the impact of bystander signaling on non‑irradiated cells, leading to elevated mutation rates even in the absence of direct dose deposition. The authors also illustrate how the model can be applied to radiotherapy scenarios, predicting tumor cell kill while accounting for normal‑tissue sparing through adaptive response mechanisms.

Limitations are openly acknowledged. The model does not incorporate spatial heterogeneity of cell populations, oxygen tension, vascular supply, or other microenvironmental factors that are known to modulate radiation response. Moreover, it lacks explicit treatment of non‑uniform exposure geometries and microdosimetric spectra, which are essential for accurately modeling high‑LET (linear energy transfer) particles or complex clinical beams. The authors suggest that future work should focus on integrating experimentally derived parameter sets, extending the stochastic tree to include tissue‑level interactions, and coupling the framework with detailed dosimetric models to enable realistic risk assessment and treatment planning.

In summary, the paper presents a novel, physics‑inspired stochastic biophysical model that treats irradiated cells as complex probabilistic systems. By merging a probability‑tree representation with quasi‑Markov chain Monte Carlo sampling and incorporating nucleation/catastrophe theory, adaptive response, and bystander effects, the authors provide a versatile tool for exploring non‑linear dose‑response relationships, radio‑adaptation, and therapeutic outcomes. The work lays a solid theoretical foundation for future extensions toward more biologically realistic and clinically applicable radiation‑biology simulations.