Mirror Oncology: A History-Dependent Mathematical Framework for Cancer Cell Adaptation and Drug Resistance — Version 2.0
Year: 2026
Type: dataset
Abstract: Cancer cells may exhibit different responses to therapeutic exposure depending not only on their current state but also on their previous treatment history. This work presents the second version of Mirror Oncology, an exploratory mathematical and computational framework designed to investigate history-dependent cancer-cell adaptation and drug resistance. The framework incorporates a dynamic memory variable into a coupled system describing drug exposure, drug-sensitive and drug-resistant cell populations, and memory-mediated population transitions. The proposed model is expressed through a system of differential equations in which the internal memory state evolves according to drug exposure and decay, potentially influencing subsequent population dynamics. Version 2.0 integrates ten computational stages, including baseline simulations, memory-state comparisons, history-dependent trajectory analysis, parameter sensitivity assessment, comparison with memoryless models, synthetic predictive experiments, robustness investigations, model-recovery studies, and exploratory transcriptomic analysis. The computational investigation examines whether systems with identical present population states but different modeled histories can generate distinct future trajectories. In addition, publicly available transcriptomic data from GEO dataset GSE210921, associated with gemcitabine-resistance acquisition in the 5637 bladder cancer cell line, are explored to characterize gene-expression differences across experimental phases. The results provide a computational demonstration of history-dependent behavior within the proposed mathematical framework. However, the model remains exploratory. Synthetic simulation results do not constitute biological validation, and transcriptomic phase differences do not directly measure the model’s memory variable or establish causal drug-resistance mechanisms. This release provides the mathematical formulation, computational implementation, simulation outputs, and exploratory analyses as a foundation for future model refinement, independent evaluation, and further investigation of history-dependent cancer dynamics.
Author: Qasemi, Marziyeh
Language: English
Subfield: Modeling and Simulation
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225454226