🧬

Biology & Life

Swarm intelligence, ant optimisation and cellular automata. Simple rules give rise to complex behaviour of living systems.

12+ simulations Three.js · Canvas 2D Boids · ACO · Conway

Category Simulations

Open a simulation — it runs right in your browser

🐦
Popular ★★☆ Moderate
3D Boids — Swarm Intelligence
5,000+ bird agents in 3D space: separation, alignment, cohesion. InstancedMesh for smooth 60fps. Adjust forces in real time.
Three.js InstancedMesh Reynolds
🐜
★☆☆ Easy
Ant Colony (ACO)
200 ants leave pheromone trails to food. The colony self-organises the shortest path — ant colony optimisation algorithm.
Canvas 2D ACO Stigmergy
🔬
★★☆ Moderate
Conway's Game of Life
Classic cellular automaton in a 3D Three.js view. Thousands of cells update synchronously under four rules. Draw your own patterns.
Three.js Cellular Automata Conway
🦠
New ★★☆ Moderate
SIR Epidemic Model
Particles spread infection by proximity. Tune transmission rate β, recovery γ, vaccinate live. Watch R₀ and the epidemic curve in real time.
Canvas 2D SIR Model Epidemiology Particles
🌱
New ★★☆ Moderate
L-Systems — Procedural Plants
Lindenmayer rewriting rules grow plants, fractals and space-filling curves. 10 presets — Koch, Hilbert, Dragon, Fern, Sierpinski and more.
Canvas 2D L-System Turtle Graphics Procedural
🐠
New ★★☆ Moderate
Prey–Predator (Lotka–Volterra)
Hundreds of prey and predator particles interact via spatial rules. Watch populations oscillate in classic Lotka–Volterra cycles.
Canvas 2D Lotka–Volterra Particles Ecology
🧠
New ★★★ Advanced
Neural Network — Backpropagation
A two-layer network learns XOR (or a 2D boundary) live. Weights shown as coloured edges, activations as node brightness, loss curve per epoch.
Canvas 2D Backprop XOR Deep Learning
🦊
New ★★☆ Moderate
Fox & Rabbits
A live Lotka-Volterra predator-prey simulation with animated animals. Watch population oscillations in a live chart!
Canvas 2D Kids Lotka–Volterra Ecology
🔵
★★☆ Moderate
Reaction–Diffusion
Gray–Scott reaction-diffusion system rendered with GPU shaders. Spot, stripe and maze patterns emerge from two interacting chemicals.
WebGL Gray–Scott Pattern
🦠
★★☆ Moderate
Disease Spread
Agent-based epidemic simulation with SEIR model. Adjust infection rate, recovery and immunity to see how diseases spread.
SEIR Canvas 2D Epidemiology
🧬
★★★ Advanced
Genetic Algorithm
A population evolves to solve the Travelling Salesman Problem. Selection, crossover and mutation drive convergence.
Canvas 2D TSP Evolutionary
🐦
★☆☆ Beginner
Boids Flocking
Craig Reynolds' three-rule flocking algorithm: separation, alignment and cohesion. Thousands of agents produce murmurations without any central control — emergent collective intelligence.
Agent-Based Emergence Three.js
🦅
★☆☆ Beginner
Bird Flock
3D murmuration of thousands of starlings. Neighbour-distance interactions only — no global leader. Watch the flock split, merge and swirl around virtual predators.
Boids Three.js Instancing
🦋
★☆☆ Beginner
Butterflies
Perlin-noise-driven butterfly flight paths in a meadow. Each butterfly has individual wing-beat frequency and attraction to flowers — a calm, meditative simulation.
Perlin Noise Three.js Animation

Learning Resources

Articles and tutorials about the algorithms in this category

Article Boids Algorithm: flocks and swarms Reynolds' three rules: separation, alignment, cohesion. FOV and spatial hashing. → Article ACO: Ant Colony Optimisation Pheromone trails. Probabilistic path selection. Solving the travelling salesman problem. → Article Conway's Game of Life Cellular automata. Birth and survival rules. Glider guns and oscillators. →
All articles →

About Biology & Life Simulations

Flocking, evolution, epidemics, and cellular patterns — modelled as code

Biology and life simulations reveal how complex behaviours emerge from simple local rules. Reynolds's three steering rules (separation, alignment, cohesion) produce realistic bird-flocking with hundreds of agents; Turing's reaction-diffusion equations generate the spot and stripe patterns seen on animal coats; the SIR epidemic model tracks disease spread through networked populations; and Conway's Game of Life demonstrates how a two-state rule set generates gliders, oscillators, and self-replicating structures.

These agent-based and equation-driven models are used in ecology, epidemiology, evolutionary biology, and synthetic biology research. By adjusting reproduction rates, infection probability, sensing radius, or diffusion coefficients you can observe population dynamics, extinction events, pattern formation, and the edge of computational universality — all within the browser.

Life simulations occupy a unique intersection of biology, mathematics, and computer science. The same reaction-diffusion equations that produce spots on a cheetah also model tumour growth and chemical oscillators. Genetic algorithms derived from evolution power industrial optimisation, machine learning, and drug discovery pipelines. Exploring these models interactively builds a deep intuition for how biological complexity arises from molecular-scale simplicity.

Key Concepts

Topics and algorithms you'll explore in this category

Conway's Game of LifeUniversal Turing-complete cellular automaton
DNA / GeneticsBase-pair encoding and mutation simulation
Turing PatternsReaction-diffusion morphogenesis
Neural FiringHodgkin-Huxley action potential model
Evolution / Genetic AlgorithmsSelection, crossover, mutation
Cellular AutomataLocal rules generating biological patterns

Frequently Asked Questions

Common questions about this simulation category

Is Conway's Game of Life really Turing-complete?
Yes — it has been proven that Game of Life patterns can simulate any computation, including a universal Turing machine, logic gates, and even a working copy of Life itself. The simulation here runs on a Canvas 2D grid with toroidal boundary conditions.
How do Turing patterns form?
Alan Turing's 1952 paper showed that two chemicals — an activator and an inhibitor — diffusing at different rates can spontaneously break symmetry and create stable spatial patterns. The reaction-diffusion simulation implements the Gray-Scott model, producing spots, stripes, and labyrinthine patterns by changing feed and kill rates.
What does the genetic algorithm simulate?
The genetic simulation evolves a population of agents using selection (fitter agents reproduce more), crossover (mixing parent genomes), and mutation (random bit flips). Over generations you can watch the population adapt to a fitness landscape in real time.

Other Categories