Drosophila Whole-Brain Model

A fruit fly's brain is smaller than a sesame seed, yet its roughly 140,000 neurons and 50 million synapses let it fly, forage, court, learn and remember. It is also the first adult brain mapped neuron by neuron and synapse by synapse. With this complete wiring diagram we can, for the first time, build an entire brain inside a computer — let signals flow along its real connections, watch the whole brain work together, and silence any single neuron or connection to ask what each part actually does. That makes the fly an ideal place to test how structure gives rise to function. Our progress in this direction is described below.

2026 bioRxiv

Connectome-constrained modeling identifies neurons and synapses that sustain spontaneous activity in Drosophila

Qianzhu Li, Wenhao Ping, Ke Zhang et al.

This study combines the FlyWire synapse-resolution connectome with brain-wide calcium recordings of spontaneous activity in head-fixed Drosophila to fit a perturbable whole-brain model. Systematic in silico perturbations identify key neurons and reciprocal synapses with excitatory partners that sustain brain-wide resting-state dynamics.

Zebrafish Whole-Brain Model

Zebrafish Whole-Brain Model

A larval zebrafish is a few millimetres long and transparent, with a brain of about 100,000 neurons. Through its skin, scientists can watch every neuron in the whole brain at once under a microscope — something almost no other vertebrate allows. Its brain shares the basic plan of our own, with regions for sensing, moving, emotion and learning. Putting its wiring, its brain-wide activity and its behavior into one model yields a digital fish that sees, swims and escapes danger — and lets us follow, end to end, how a vertebrate brain turns what it sees into what it does. Our progress in this direction is described below.

Human Whole-Brain Model

Human Whole-Brain Model

The human brain has some 86 billion neurons, wired by white-matter fibres into a network that spans the whole brain. No two people's networks are quite the same — perhaps one reason we each think, feel and react differently. MRI can map how each person's brain regions are connected; EEG and MEG record its activity millisecond by millisecond. A human whole-brain model brings the two together into a digital brain for each individual — one that reproduces brain activity and also generates behavior — letting us explore, without touching a real brain, where individual differences come from, how brain disorders arise, and which interventions might work best. Our progress in this direction is described below.

2026 bioRxiv

A brain-behavior generative model links individual brain dynamics to perceptual decisions

Chaoming Wang, Guochun Yang, Shaonan Wang et al.

This study introduces a brain-behavior generative model in which whole-brain dynamics on each participant's structural connectome generate both single-trial EEG and reaction times. Fitted to 17 participants performing motion discrimination, the models reproduce individual event-related potentials and reaction-time distributions and predict whether held-out trials are relatively fast or slow. In silico silencing predicts where, how, and when each person's brain controls decision speed: sensorimotor regions act consistently across participants, whereas the same parietal region has opposite effects in different individuals.