(Paris) An international team has succeeded in mapping all the neurons and connections in the brain of the Drosophila fly, a major advance in neuroscience, and its results were published on Wednesday in the journal Nature.
“One in three people in the world lives with a neurological or neuropsychiatric disorder […] Most of them are related to “circulatory disorders, that is, something is not working right in the way neurons communicate with each other in the brain,” recalls John Ngai, director of the Brain Initiative at the US Institutes of Health.
However, “we know too little about the circuits of the human brain to be able to offer treatment,” he continued during a press conference in which he presented the complete “neural network” of the fruit fly, a key step in understanding these mechanisms.
A connectome is a representation of neural connections in the brain, which describes how neurons interact with each other to form brain circuits. The term refers to the genome, the genetic material of an organism, the first decodings of which revolutionized science and medicine.
Google Maps for the brain
It took hundreds of researchers from the Flywire Consortium ten years to map the 140,000 neurons and 50 million synapses in the fruit fly brain.
This little fly, which flies around the fruit in our kitchens, is an important model for neuroscience because its brain solves many problems similar to ours.
It is capable of sophisticated behaviors such as walking, flight, learning, memory, navigation, and even social interactions.
The fruit fly's brain was no larger than a grain of sand, and was first cut into 7,000 sections, then imaged using high-resolution microscopes and assembled to form a 3D image of the brain.
“We started by manually identifying neurons, but we estimated that it would take more than 4,000 years of human work to produce a connectome,” says Gregory Jeffress, of the University of Cambridge's Laboratory of Molecular Biology.
The consortium then called for artificial intelligence. “But the AI made errors in reconstructing neurons and had to be corrected manually,” says Mala Murthy of Princeton University, co-founder of the consortium, which “opened up the data to neuroscientists across the entire community working on different parts of the fly brain to… They can help us.”
Once the map was created, neuroscientists named the different types of neurons and connections. “It's a bit like Google Maps, but for the mind,” Mr Jeffress explains.
“The diagram of elementary connections is a bit like a satellite image of the Earth, which shows streets, buildings and rivers. Explaining the neurons means putting together the names of streets, towns, opening hours…” explains the researcher. “You need both the base map and these annotations for it to be really useful to scientists.”
Neuroscientists have identified more than 8,000 cell types, making it the largest atlas ever produced. They also used artificial intelligence to predict the neurotransmitter of each neuron, and whether the connections were excitatory or inhibitory.
This work has already enabled many advances in understanding the fly's visual, olfactory and motor functions.
But if this neural network represents a “critical milestone,” the task remains “huge” in understanding how the brain works, warns Mr. Ngai.
“The function of most cell types in the fly brain is still unknown,” says Sebastian Seung, a neuroscientist at Princeton University. Sebastian Seung highlights the need for “other types of measurements in living brains to understand how these complex circuits underlie behavior.”
Just as sequencing the genomes of small organisms led to the rapid development of sequencing mammalian genomes, the Drosophila connectome will serve as the basis for mapping more complex brains.
The human brain's neural network – about a million times more complex than that of a fruit fly, containing 86 billion neurons – is currently elusive.
The next step is to map the mouse brain, which the researchers hope to create within five to ten years.
