Mathematical Biology

Mathematical biology is the branch of applied mathematics that turns biological questions into mathematical ones and uses the answers to understand living systems.

12 frameworks, from 1838 to today · Applied Mathematics · Mathematics

still a working view ended · each leads to its entry below

Overview

Mathematical biology is the branch of applied mathematics that turns biological questions into mathematical ones and uses the answers to understand living systems1. Its practitioners build models—simplified, often idealized descriptions of how populations, cells, epidemics, or molecules behave—and then analyze them to predict, explain, or design experiments1. What is at stake is not merely mathematical elegance but biological insight: whether a model captures why a population cycles, how a pattern of stripes forms on an animal's skin, or whether a drug will control a tumor. The field's history is one of repeated crossings between mathematics and the life sciences, each crossing driven by a biological puzzle that existing mathematics could not handle.

Populations, Epidemics, and the First Mathematical Models

The earliest sustained mathematical biology grew out of demography and epidemiology in the eighteenth and nineteenth centuries. In 1760, Daniel Bernoulli presented to the French Academy a mathematical analysis of smallpox inoculation, modeling how variolation would change mortality across an age-structured population. His work was among the first to treat a public-health intervention as a mathematical problem, though it did not immediately found a tradition.

The decisive turn came with the work of Pierre-François Verhulst, who in 1838 proposed the logistic equation for population growth: a population grows exponentially at low density but slows as it approaches a carrying capacity set by resources. Verhulst's curve was largely forgotten until the twentieth century, when Raymond Pearl and Lowell Reed rediscovered and promoted it in the 1920s as a description of human and laboratory populations. The logistic model's core idea—that growth rates depend on density—became a foundation of population dynamics.

Epidemic modeling took shape in the same era. In 1927, William Kermack and Anderson McKendrick published a compartmental model of epidemics, dividing a population into susceptible, infected, and recovered groups (the SIR scheme) and deriving a threshold condition for an outbreak to spread. Their threshold theorem—that an epidemic grows only if the density of susceptibles exceeds a critical value—remained the conceptual core of epidemic theory. Compartmental modeling, in which individuals are assigned to states and flow between them according to rate equations, became a general tool for infectious disease, and later for physiology and pharmacology2.

Ecology, Genetics, and the Mid-Century Synthesis

The 1920s and 1930s saw mathematical population biology mature on two fronts. In ecology, Alfred Lotka and Vito Volterra independently developed predator–prey equations showing that interacting populations could oscillate indefinitely rather than settle to equilibrium. Lotka's 1925 Elements of Physical Biology and Volterra's 1926 analysis of fish populations in the Adriatic established the Lotka–Volterra framework, which became the template for modeling competition, predation, and symbiosis. The key insight was that simple interaction rules could generate complex dynamics—cycles, coexistence, or extinction—without external forcing.

In genetics, the same decades produced population genetics3. Ronald Fisher, J. B. S. Haldane, and Sewall Wright independently showed that Mendelian inheritance combined with natural selection could explain evolutionary change quantitatively. Fisher's 1930 The Genetical Theory of Natural Selection used diffusion approximations to model how allele frequencies change under selection, mutation, and drift in large populations. Wright introduced the concept of adaptive landscapes and the shifting-balance theory of evolution, while Haldane worked out selection coefficients for real traits such as industrial melanism in moths3. Their mathematical framework—allele frequencies, fitness coefficients, Hardy–Weinberg equilibrium—became the language of evolutionary biology34. The field's central equation, the change in allele frequency under selection, is a simple difference equation whose analysis reveals when selection maintains variation and when it eliminates it.

Excitable Media and the Biophysics of Nerve

A different strand emerged from physiology. The nerve impulse posed a problem that nineteenth-century physics could not solve: how does a signal propagate along a fiber without decaying? The answer came from biophysics. In the 1950s, Alan Hodgkin and Andrew Huxley, building on earlier work by Kenneth Cole and others, measured ionic currents in the squid giant axon and formulated a system of nonlinear differential equations describing how membrane voltage depends on sodium and potassium conductances. Their 1952 model, which won the Nobel Prize in 1963, was a landmark of mathematical biology: a set of ordinary differential equations, fitted to experimental data, that reproduced the shape and speed of the action potential5.

The Hodgkin–Huxley equations also inspired a broader mathematical program. Excitable media—systems that rest quietly but fire when perturbed—became a general class. In the 1960s and 1970s, mathematicians studied simplified models such as the FitzHugh–Nagumo equations, which reduce the Hodgkin–Huxley system to two variables and make the geometry of excitation visible6. The traveling-wave solutions of these equations explained how impulses propagate, and the same mathematics was later applied to cardiac tissue, slime molds, and chemical oscillators.

Reaction–Diffusion and the Mathematics of Pattern

In 1952, the same year as Hodgkin and Huxley's paper, Alan Turing published "The Chemical Basis of Morphogenesis," proposing that interacting chemical substances diffusing at different rates could spontaneously form spatial patterns—spots, stripes, and waves—from an initially uniform state78. Turing's reaction–diffusion model showed that a system stable without diffusion could become unstable when diffusion was added, a counterintuitive result that became known as diffusion-driven instability8. The idea was that biological pattern formation, such as the arrangement of digits or the markings on animal coats, might arise from such chemical prepatterns.

Turing's paper was largely ignored by biologists at the time, but it seeded a mathematical tradition. In the 1970s, researchers such as Hans Meinhardt and Alfred Gierer developed specific reaction–diffusion models for biological patterning, and the mathematics of pattern formation became a recognized subfield. The 1990s brought experimental support when scientists identified candidate molecular mechanisms for Turing patterns in fish skin and other systems, though the debate over how widespread such mechanisms are continues.

Stochasticity, Networks, and the Computational Turn

Through the mid-twentieth century, most mathematical biology was deterministic: given initial conditions, the future was fixed. But biological systems are noisy, and small numbers of molecules matter. In the 1940s, while working on the Manhattan Project, Nicolas Metropolis and Stanislaw Ulam developed the Monte Carlo method, and by the 1970s stochastic simulation of chemical kinetics—the Gillespie algorithm, published by Daniel Gillespie in 1977—made it practical to simulate systems where random fluctuations drive outcomes. Stochastic modeling became essential for gene expression, where a single cell may contain only a few copies of a regulatory molecule, and for population genetics, where drift matters in small populations.

The late twentieth century also brought network thinking. Systems biology, emerging in the 1990s and 2000s, treated cells as networks of interacting genes, proteins, and metabolites rather than as collections of independent pathways. Mathematical tools from graph theory, dynamical systems, and control theory were applied to understand robustness, feedback, and modularity in these networks. The goal was often to explain why biological networks have the properties they do—why they tolerate perturbation, how they switch between states—rather than to model any single component in detail.

Agent-based and individual-based modeling grew alongside this computational turn. Instead of treating populations as continuous densities, these models simulate many discrete individuals, each following simple rules, and observe the collective behavior that emerges9. The approach was used for ecosystems, immune systems, and tumor growth, and it became feasible as computing power grew in the 1990s and 2000s. Its appeal was that it could capture heterogeneity and local interactions that aggregate models smoothed away.

Calibration, Oncology, and the Data-Rich Present

As models grew more complex, the problem of fitting them to data became central. Computational and data-driven model calibration—using optimization, Bayesian inference, and machine learning to estimate parameters and quantify uncertainty—became a subfield in its own right. The rise of high-throughput experimental data in the 2000s and 2010s made this both possible and necessary: a model with dozens of parameters cannot be fitted by hand, and the question of which parameter sets are consistent with observations became a mathematical problem in itself.

Mathematical oncology emerged as a distinct application in the 1990s and 2000s, applying population dynamics, reaction–diffusion, and stochastic models to tumor growth, invasion, and treatment response. The field's aim was to understand why tumors grow the way they do and to design better treatment schedules—for example, using models to predict when resistance will emerge under different drug regimens. It drew on the older traditions of population modeling and pattern formation, applying them to a clinical problem.

The Present State

Mathematical biology today is pluralistic. Population dynamics and ecological modeling remain active, with attention to stochasticity, spatial structure, and evolution. Compartmental epidemic models, descendants of Kermack and McKendrick, became globally familiar during the COVID-19 pandemic, when their predictions and limitations were debated in public. Population genetics has expanded into genomics, where mathematical models of selection, drift, and recombination are used to interpret large-scale sequence data. Reaction–diffusion and excitable-media models continue to explain pattern and propagation in development and physiology. Network and systems biology modeling has become central to interpreting molecular data, and agent-based models are used wherever individual heterogeneity matters. Computational calibration and machine learning are now routine parts of the modeling workflow.

The open questions are as much biological as mathematical: how to build models that are both simple enough to analyze and complex enough to be useful; how to connect models across scales, from molecules to cells to populations; how to quantify uncertainty when data are sparse; and how to use models not just to predict but to design experiments and interventions. The field's history suggests that its most durable contributions come when a biological puzzle forces a new mathematical idea—and when that idea, once found, turns out to illuminate problems far beyond its original home.

References

  1. Mathematical and theoretical biology, Wikipedia
  2. Compartmental models (epidemiology), Wikipedia
  3. Population genetics (introduction, History), Wikipedia
  4. Hardy–Weinberg principle, Wikipedia
  5. Hodgkin–Huxley model, Wikipedia
  6. FitzHugh–Nagumo model (introduction, History), Wikipedia
  7. The Chemical Basis of Morphogenesis, Wikipedia
  8. Turing pattern, Wikipedia
  9. Agent-based model, Wikipedia

The frameworks of Mathematical Biology

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1838-Present
  2. 1925-Present
  3. 1927-Present
  4. 1930-Present
  5. 1945-Present
  6. 1952-Present
  7. 1952-Present
  8. 1961-Present
  9. 1990-Present
  10. 1990-Present
  11. 1995-Present
  12. 2000-Present
A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Mathematical Biology is a planet in Nousfarer

Its city has a tower for each of these 12 frameworks. Land, walk in, and climb one concept at a time.

Land on Mathematical Biology

Related fields

More in Applied Mathematics

See where Mathematical Biology sits on the Map of Human Knowledge