How Scientists Mapped Every Neuron in a Male Fruit Fly’s Central Nervous System

A complete connectome is more than a picture of a brain: it is a cellular wiring diagram showing how neurons are connected. The new adult male Drosophila resource maps the central brain, optic lobes, and ventral nerve cord of one animal as a continuous system. It contains 166,691 fully proofread neurons, 11,691 annotated cell types, and 125 million synaptic connections. 25

The breakthrough: a whole connected nervous system

Previous fly datasets established that whole-brain connectomics was possible. The published adult female brain connectome, for example, reconstructed 139,255 neurons and roughly 50 million chemical synapses. 7

The male dataset expands the scope in two important ways. First, it is larger by neuron count. Second, it includes the ventral nerve cord—the fly structure often compared with a spinal cord—while preserving the neck connection to the brain. That means researchers can trace circuits from sensory processing and higher brain centers into pathways involved in movement and behavior rather than analyzing the brain and nerve cord as disconnected maps. 2

How the team made an enormous reconstruction usable

This was the culmination of nearly two decades of work by Janelia and collaborators, including researchers at Google Research, the MRC Laboratory of Molecular Biology, and the University of Cambridge.

The project did not hinge on a single technique. It required an integrated workflow:

• Electron microscopy generated images detailed enough to resolve neurons and their synaptic connections. The FlyEM program uses EM as the foundation for mapping neurons and synapses across the Drosophila nervous system. 8
• AI-assisted computation helped turn image data into candidate cellular reconstructions and supported work at a scale that manual tracing alone could not match. Google describes the collaboration as a decade-long partnership using computing and AI to build cellular-scale maps of whole brains.
• Human proofreading and annotation converted computational output into a finished scientific resource. The final dataset is described as fully proofread and comprehensively annotated, a critical distinction from an unverified automated reconstruction. 25
• Cell-type and molecular labeling made the wiring diagram interpretable. The atlas includes annotations related to the sex-determination genes fruitless and doublesex. 5

The practical lesson is that AI made the scale tractable, but expert review made the map reliable enough to use for biological questions.

What male–female comparison can now reveal

A complete male CNS map alongside the prior female brain connectome makes direct comparisons possible at synaptic resolution. The male-connectome study cross-matched 7,319 cell types in the central brain and reported 7,205 isomorphic types, 114 dimorphic types, 262 male-specific types, and 69 female-specific types. 5

That comparison gives researchers a way to distinguish between two possibilities that behavior studies alone often cannot resolve: a behavior may arise from a neuron type found only in one sex, or from a neuron type present in both sexes but connected differently. The fruitless and doublesex annotations provide an additional molecular reference point for investigating sexually dimorphic circuits. 56

From taste to feeding decisions

The larger value of a whole-CNS map is that it can be paired with functional experiments. Work combining the whole-brain connectome with calcium imaging found that many early taste-responsive neurons in the subesophageal zone encode a single taste modality, including local interneurons and projection neurons.

Related connectomic analysis found that second-order taste neurons are largely segregated by taste modality in this region. Third-order neurons project more broadly outside it and show more overlap between modalities, an organization that could help transform taste signals into feeding-related decisions.

A companion report also describes tracing taste signals through to motor pathways involved in feeding and identifies routes that could explain anticipatory insulin release before nutrients are absorbed. The supplied reporting supports this as a potential circuit-level explanation; it does not establish a definitive mechanism for insulin signaling.

What the map does—and does not—mean for bigger brains

The male fly connectome offers a practical blueprint for connectomics: high-resolution imaging, scalable automated reconstruction, rigorous proofreading, standardized annotations, and tools that let other scientists query and reuse the data. The resource is available for interactive exploration and programmatic access. 1

It does not make a complete mouse or human connectome imminent. Larger brains present much steeper problems in tissue volume, imaging, reconstruction, storage, computation, and quality control. But the fly work provides a demanding real-world test of the pipeline needed for future projects, including systems where anatomy, neural activity, and behavior can be studied together.

For neuroscience, the lasting advance is not merely a record-sized neuron count. It is a proofread, searchable map that connects brain and nerve cord in one animal—and gives researchers a common reference for testing how neural wiring produces sex-specific behavior, sensory decisions, and movement. 25