Biological Mathematics

Biological Mathematics applies mechanistic and quantitative tools to understand life at scales ranging from molecules to ecosystems. Core approaches include differential equations to describe reaction–diffusion processes in cells and tissues, stochastic models to capture demographic variability, network theory for gene and protein interactions, and agent-based simulations to resolve individual behaviours in populations. Recent advances harness machine learning to estimate model parameters and integrate data-driven insights with first-principles frameworks. Digital twins—personalised computational replicas of biological systems—enable in silico experimentation under uncertainty, informing precision medicine and adaptive management in ecology. Together, these methods seek to reveal general principles—such as pattern formation, homeostasis and evolutionary trade-offs—by balancing mathematical rigour with biological realism.

Research from Nature Portfolio

Studies have revealed how subcellular structures govern collective cell migration modes. One work correlated three-dimensional lamellipodium morphology in keratocytes with intracellular diffusion dynamics, demonstrating a reversible switch between fast and slow migration. Deformation of the lamellipodium front alters molecular crowding, thereby tuning speed and steering. A minimal physical model then represented crawling cells as droplets of active polar fluid with localised treadmilling and contractility, reproducing diverse motility regimes without explicit regulatory input.
Another investigation employed multiparametric MRI to personalise biomechanical models of high-grade glioma response to chemoradiation. By integrating diffusion-weighted and contrast-enhanced images into a two-species mathematical framework, the study captured tumour cell density heterogeneity and forecast spatial response patterns at follow-up visits. A model selection procedure identified a parsimonious description of enhancing and non-enhancing compartments that achieved low prediction error in individual patients, illustrating the promise of image-driven model calibration for treatment planning.

Research from all publishers

A predictive digital twin framework has been developed to optimise radiotherapy schedules in high-grade gliomas under biological and measurement uncertainties. The twin is initialised through population priors for mechanistic growth parameters and personalised via Bayesian model calibration using serial MRI data. Multi-objective, risk-based optimisation then generates patient-specific regimens balancing tumour control probability against normal tissue toxicity, yielding median gains in time to progression and potential radiation dose reductions compared with standard protocols.
In tumour spheroid biology, time-varying oxygen conditions were imposed in vitro to mimic hypoxic cycling. Coupling experimental data with a reaction–diffusion–consumption model revealed unexpected adaptive behaviours, including transient necrotic core clearance and reversal of growth phase dynamics. These phenomena underscore the importance of microenvironmental fluctuations in spheroid evolution and model validation.
In ecological modelling, an intuitionistic fuzzy predator–prey system with Holling type II functional response addressed uncertainty in toxin exposure and harvest delays. By representing model parameters as fuzzy sets and applying intuitionistic fuzzy operators, researchers derived local and global stability criteria in the presence of size-selective harvesting and multi-toxin contamination. Numerical examples illustrated how imprecision in toxicant levels and time lags can critically affect population persistence and yield management.

Biological Mathematics publication trend

The graph below shows the total number of articles in biological mathematics across all publications each year (not limited to Nature Index journals).

Technical terms

Reaction–diffusion equation: A partial differential equation coupling local reaction kinetics with spatial diffusion to describe concentrations over time and space.

Agent-based model: A computational framework in which individual entities follow prescribed rules, enabling emergent collective behaviours to be simulated.

Digital twin: A patient- or system-specific computational replica calibrated to data, used to test and optimise interventions under uncertainty.

Active polar fluid: A continuum model treating cells as fluid droplets with internal oriented stresses representing cytoskeletal activity.

Intuitionistic fuzzy set: A generalisation of fuzzy sets that introduces degrees of membership, non-membership and hesitation to handle parameter uncertainty.

Basic reproduction number (R₀): The average number of secondary cases generated by one infected individual in a fully susceptible population, governing epidemic thresholds.

References

1. Switch of cell migration modes orchestrated by changes of three-dimensional lamellipodium structure and intracellular diffusion. Nature Communications (2023).
2. A minimal physical model captures the shapes of crawling cells. Nature Communications (2015).
3. Image-based personalization of computational models for predicting response of high-grade glioma to chemoradiation. Scientific Reports (2021).
4. Predictive digital twin for optimizing patient-specific radiotherapy regimens under uncertainty in high-grade gliomas. Frontiers in Artificial Intelligence (2023).
5. Growth and adaptation mechanisms of tumour spheroids with time-dependent oxygen availability. PLOS Computational Biology (2023).