🧠

Neuroscience & Biophysics

From neural networks to membrane transport — explore the biophysics of cells, neurons, and physiological systems interactively.

8 simulations Canvas 2D · WebGL Uses Backpropagation, FitzHugh-Nagumo, Monte Carlo

Neuroscience Simulations

Open a simulation — it runs right in your browser, no installation needed

🧠
Popular ★★☆ Moderate
Neural Network
Interactive backpropagation visualiser. Build a multi-layer perceptron, pick a dataset, train the network and watch weights, biases, and the decision boundary evolve in real time.
Canvas 2D Backprop Perceptron
❤️
★★★ Advanced
Cardiac Action Potential
FitzHugh-Nagumo excitable-media model on a 2D grid. Watch spiral-wave re-entry, pacemaker foci, and conduction blocks — the electrophysiology behind cardiac arrhythmias.
Canvas 2D FitzHugh-Nagumo Excitable Media
🔬
★☆☆ Beginner
Cell Membrane Diffusion
Passive, facilitated, and active transport across a cell membrane. Adjust concentrations, channel density, and ATP availability to maintain cellular homeostasis.
Canvas 2D Cell Biology Transport
🧬
★★☆ Moderate
Protein Folding
HP lattice model Monte Carlo protein folding. Watch a chain of hydrophobic (H) and polar (P) residues explore conformational space, seeking the lowest energy state.
Canvas 2D Monte Carlo HP Model
🩸
★★☆ Moderate
Blood Flow
Poiseuille flow in a cylindrical vessel. Visualise velocity profiles, wall shear stress, and flow rate. Non-Newtonian Casson model for realistic blood viscosity.
Canvas 2D Poiseuille Navier-Stokes
🌀
★★☆ Moderate
Reaction-Diffusion
Turing pattern formation via Gray-Scott reaction-diffusion. Adjust feed and kill rates to produce spots, stripes, labyrinthine patterns and solitons.
Canvas 2D Gray-Scott Turing Patterns
🧠
New ★★☆ Moderate
Brainwave Oscillations
Kuramoto coupled-oscillator EEG model with four frequency bands: delta (δ), theta (θ), alpha (α) and beta (β). Adjust brain-state preset and coupling strength to watch synchrony emerge.
Canvas 2D Kuramoto EEG
🧠
New ★★★ Advanced
Spiking Neural Network
Network of 50 LIF neurons (40 excitatory + 10 inhibitory) with STDP synaptic plasticity. Scrolling raster plot, membrane voltage trace and real-time firing-rate statistics.
Canvas 2D LIF STDP Raster Plot
New ★★★ Advanced
Hodgkin–Huxley Neuron
Conductance-based action potential model (Nobel 1963). Inject current and watch Na⁺ and K⁺ gating variables m, h, n drive the spike, refractory period and repetitive firing.
ODE Solver Ion Channels Action Potential
🧠
New ★★ Intermediate
Synaptic Transmission
Calcium-triggered vesicle release, neurotransmitter diffusion across the synaptic cleft, receptor binding and EPSP/IPSP generation. Switch between Glutamate and GABA to see excitatory and inhibitory responses.
Synapse Neurotransmitter EPSP / IPSP
🔄
New ★★☆ Moderate
Kuramoto Synchronisation
N coupled phase oscillators with Gaussian natural-frequency spread. Adjust coupling K and watch the order parameter r(t) jump from incoherence to full synchrony at the critical threshold K_c = 2σ.
Coupled Oscillators Order Parameter Phase Transition

About Neuroscience & Biophysics Simulations

Neurons, membranes, proteins — the machinery of life, visualised

Neuroscience and biophysics simulations bring the molecular and cellular machinery of life into interactive focus. From the cascading ion channels of a cardiac action potential to the folding pathways of a protein chain, these models let you manipulate parameters and observe emergent biological behaviour in real time.

The neural network visualiser implements full backpropagation learning — watch weights adjust as the network learns to classify data. The FitzHugh-Nagumo model captures the excitable membrane dynamics that underpin every heartbeat and nerve impulse, while membrane diffusion demonstrates how passive, facilitated, and active transport maintain cellular homeostasis.

These simulations connect abstract equations to tangible biology. The same Navier-Stokes simplifications used in the blood flow model inform real medical device design, and Monte Carlo methods for protein folding drive modern drug discovery pipelines. Running them in a browser makes computational biophysics accessible without specialised software.

Key Concepts

Topics and algorithms you'll explore in this category

BackpropagationGradient descent for training multi-layer neural networks
FitzHugh-NagumoSimplified excitable membrane model for action potentials
Membrane TransportPassive, facilitated, and active ion channel mechanisms
Protein FoldingHP lattice model with Monte Carlo conformational search
Poiseuille FlowLaminar flow in cylindrical vessels under pressure gradient
Reaction-DiffusionPattern formation through chemical morphogen interaction

🧠 Test Your Neuroscience Knowledge

Five quick questions to check your understanding of the brain and nervous system

Neuroscience Quiz

Frequently Asked Questions

Common questions about this simulation category

How does the neural network learn?
The network uses backpropagation — computing the gradient of the loss function with respect to each weight via the chain rule, then updating weights in the direction that reduces error. You can watch the decision boundary evolve as the network trains on your data.
What is the FitzHugh-Nagumo model?
It is a two-variable simplification of the Hodgkin-Huxley equations that captures the essential dynamics of excitable cells — a fast activation variable (membrane voltage analogue) and a slow recovery variable. It produces action-potential-like spikes when stimulated above threshold.
How is blood flow simulated?
The simulation uses Poiseuille's law for laminar flow in cylindrical vessels. Pressure gradient, vessel radius, and blood viscosity (non-Newtonian via the Casson model) determine flow rate and wall shear stress — the same physics used in vascular surgery planning.

Other Categories