Researchers have completed a full wiring diagram of the adult fruit fly central nervous system, mapping more than 166,000 neurons and their connections across the brain and ventral nerve cord. Published in Nature on October 2, 2024, the milestone achievement scales up connectomics research to help scientists decode how brains generate complex behavior.
Two Decades of Scale: From Worms to Flies
When researchers at the Howard Hughes Medical Institute’s Janelia Research Campus set out to create a comprehensive wiring diagram of all neurons in the fruit fly brain in 2008, skeptics thought the project was nearly impossible. It had been 20 years since scientists mapped the 302 neurons of the tiny worm C. elegans, an effort that took more than a decade. Most scientists argued that mapping the fly brain, which contains over 100,000 neurons, would take too long, cost too much, and yield limited insight into actual brain function.
Those doubts failed to deter Janelia founding Executive Director Gerry Rubin, who bet that improved imaging and computational techniques could slash both time and expense. The resulting dataset maps the central nervous system of a fruit fly, capturing more than 166,000 neurons across the brain and ventral nerve cord.
The Anatomy of the FlyWire Connectome
The complete wiring map, announced in a nine-paper package in Nature, documents nearly 140,000 neurons in the brain alone, alongside more than 50 million synapses. About 85% of these neurons are intrinsic to the brain, meaning they synapse exclusively with other brain cells, allowing the organ to communicate primarily with itself. These intrinsic neurons vary widely in size, stretching from less than 0.2 millimeters to nearly 20 millimeters in length.
Tools such as flood-filling networks and the PATHFINDER reconstruction system helped turn flat images into accurate 3D shapes. Researchers then classified and annotated more than 8,400 distinct cell types based on their location, connectivity, developmental origin, and shape.
The resulting atlas serves as a navigational guide for neurobiology.
Mapping the Subesophageal Zone and Information Flow
Because individual neurons form synapses in only a few regions while connecting across many others, the team computed a projectome to track links between different parts of the brain. This projection map uncovered crucial details about the subesophageal zone, a region historically underrepresented in earlier wiring diagrams. The analysis shows that the subesophageal zone interacts with virtually all parts of the brain, receiving a large fraction of incoming signals and dispatching outgoing instructions, including nearly all signals sent to motor neurons.
With this detailed connectivity chart, scientists can trace information flow from sensory input to physical action. Researchers mapped the ocellar circuit, which helps orient the fly’s body during flight in response to visual stimuli, illuminating the precise mechanical path of orientation behavior.
Simulating an Entire Brain on a Laptop
Having a complete wiring map allowed scientists to test whether static anatomical data could actually predict living neural activity. Phil Shiu, a former postdoctoral fellow at the University of California, Berkeley, took the 139,255 neurons and 50 million connections and built a computer model capable of running entirely on a laptop.
When the model simulated sugar-sensing neurons detecting sugar water, it successfully predicted the neural cascade that extends the insect’s proboscis to feed. When bitter compounds were introduced alongside sweet ones, inhibitory neurons in the simulation stopped the process to prevent poisoning. Other researchers used the model to accurately predict leg-grooming behavior triggered by antennal dirt and to forecast locomotion patterns.
Next Steps: Toward Vertebrate Brains and Artificial Intelligence
The successful fly brain simulation points toward larger modeling ambitions. Researchers are already applying the methodological lessons learned from Janelia’s decades-long effort to transparent vertebrate fish models, specifically the larval zebrafish and adult Danionella. The long-term scientific horizon points toward mapping mouse brains and, ultimately, the human brain.
Beyond biological discovery, the computational architecture inspired by real neural wiring offers an alternative path for machine learning. By reducing the need for manual error correction through AI tools, labs worldwide can now approach larger vertebrate connectome projects within sustainable budgets and timelines.
