HHMI Janelia Research Campus, Google Research and collaborators report the first complete map of the brain and central nervous system of an adult male fruit fly. The reconstruction contains more than 166,000 neurons, classified into 11,691 neuron-cell types, and uses serial imaging, computational reconstruction and artificial intelligence to represent the nervous system in three dimensions.

The work is a large-scale example of connectomics—the reconstruction and analysis of connections between brain cells. The researchers’ visualisations show how sensory neurons can connect through intermediate neurons to motor neurons, and how some neural connections differ between male and female flies.

The source announcement describes the map as complete, but it does not provide the underlying dataset identifier, detailed methods, reconstruction error rate, imaging resolution or a quantified total number of connections. The map is also an anatomical reconstruction: it does not, by itself, establish the activity, strength, timing or causal behavioural role of every connection.

Contents

What the map contains

The mapped central nervous system includes the fly’s brain and ventral nerve cord. The ventral nerve cord is the main nerve cord of an insect and serves a role broadly analogous to the spinal cord in vertebrates.

According to the project description, the reconstruction includes more than 166,000 neurons. Computing and AI helped human experts classify those neurons into 11,691 cell types. In this context, a cell type refers to a group of neurons identified as belonging to the same recognised category in the reconstruction; the announcement does not provide the full classification criteria.

The project builds on an earlier complete map of a female fruit-fly brain. Mapping both sexes allows researchers to compare not only which neuron types are present, but also how their connections are organised.

The researchers describe three broad patterns in male-female comparisons:

  • neuron types that are present in both sexes and have comparable organisation;
  • sex-specific neurons found in only one sex; and
  • dimorphic neurons, present in both sexes but connected to different neighbouring neurons.

One example is the AOTU012 neuron type, which the visualisation shows in both male and female flies. Its neighbouring connections include some that are similar between the sexes and others that differ.

How the nervous system was reconstructed

The map was produced from thin sections of the fly’s brain and body. These sections were imaged, generating millions of two-dimensional images. Computers and AI were then used to combine those images and reconstruct three-dimensional neural shapes and connections.

This process is more than simply photographing a brain. Neurons branch extensively, and their processes can pass through many tissue sections. A connectomics reconstruction therefore attempts to follow those structures through the imaged material and identify where they connect with other neurons.

AI assisted the reconstruction and classification process, but the source describes human experts as part of the classification work. The announcement does not specify how much of the reconstruction was automated, the criteria used to determine that a connection was present, or the error rate of the resulting map.

The project’s stated outcome is a map of the complete male fruit-fly brain and central nervous system. However, the supplied source does not say whether the reconstruction represents one individual fly or multiple specimens, and it does not describe the number of tissue sections or the imaging resolution.

What the visualisations show

One of the mapped circuits follows a route from visual neurons to motor neurons. The example begins with R1-R6 visual neurons and ends at the DNg13 motor neuron. Such a pathway illustrates how sensory information can be structurally linked to neurons involved in movement.

The visualisation also identifies the male-specific neuron type LoVP92 as an intermediate step in this pathway. The project associates LoVP92 with the fly’s so-called love spot and with courtship behaviour. That association is presented by the project; the supplied source does not provide experimental effect estimates or a detailed causal test of LoVP92’s role.

The male-female comparisons provide another use for the map. A difference between two connectomes may arise because a neuron exists only in one sex, because the same neuron type has different neighbouring connections, or because a shared circuit is organised differently. These distinctions can help researchers investigate how neural wiring relates to sex-specific behaviours, although a wiring difference alone does not establish how the behaviour is generated.

Three companion studies were released alongside the map, applying it to visual systems, taste and social behaviour. The supplied announcement does not provide their study designs, results, publication venues or effect estimates, so their findings cannot be assessed in detail here.

What the map does and does not establish

A complete structural map can provide a framework for asking how sensory information is transformed into action. Researchers can use it to trace routes between sensory, intermediate and motor neurons, compare circuits between sexes and identify candidate pathways for experimental testing.

It does not show the full physiological operation of the nervous system. A structural connection does not by itself reveal whether the connection is active, how strong it is, when it operates, or whether it is necessary or sufficient for a particular behaviour. Those questions require additional measurements and experiments.

The findings also concern an adult male fruit fly. They should not be presented as a map of the human brain or as direct evidence about human behaviour. The source does not establish how directly the organisation of this nervous system generalises to other insects or vertebrates.

Finally, the announcement describes the work as a complete map but does not provide independent validation details, a quantified connection count, reconstruction error rate or the publication status of the underlying research. The map is therefore best understood as a substantial reported connectomics resource and a basis for further investigation, rather than a complete functional explanation of the fly’s behaviour.

Sources