A revolutionary map of the fly brain could change how we study our brains
Researchers have developed a groundbreaking new resource—the, described today in the journal —that maps every neuron and synaptic connection in the central brain of Drosophila melanogaster, or the fruit fly. Totaling over 130,000 neurons and 30 million synaptic connections, this revolutionary tool will expedite inquiry into how the brain works and expand the questions that can be asked.
“The importance of this cannot be understated, because it really just drastically changes the field,” said Gabriella Sterne, PhD, assistant professor of Biomedical Genetics and Neuroscience at the Del Monte Institute for Neuroscience at the University of Rochester, who contributed to this research as a member of the FlyWire consortium, a group co-led by the MRC Laboratory of Molecular Biology in Cambridge, United Kingdom, Princeton University, the University of Vermont, and the University of Cambridge. “The first time I saw the complexity of the connectome it literally blew my mind because we have been thinking of these circuits in a simplistic manner, but we can now appreciate that they are far more complex than we imagined.”
Researchers will be able to use this resource to untangle complex brain connections and functions, accelerate findings, inform machine learning and artificial intelligence, and improve our understanding of the human brain. “The connectome makes it easier to uncover general and fundamental principles that govern neural circuit function. Discovering such principles in a relatively simple brain will inform the search for similar processes in the human brain, potentially leading to unifying theories of brain function,” Sterne explained. “Once we understand the computations that neural circuits are performing in a healthy brain, we can start to ask how circuit function is disrupted in disease.”
Creating a tool to use the abundance of information
Sterne was also part of a team at University of California, Berkeley who took this resource and tested whether it could be used to understand how neural activity flows through networks of neurons that are connected by synapses.
“The connectome is really difficult to interpret,” Sterne explains. She co-authored another paper about the computational model she helped create to take on this hypothesis. “This paper showcases what the connectome can be used for – it allows researchers to move beyond single cell types and towards studying how neural circuits are organized and how groups of neurons are working together to produce brain function.”
Sterne and Philip Shiu, PhD, first author of the study, were both postdoctoral fellows in the lab of Kristin Scott, PhD, who led this research, when Shiu began to wonder whether brain function could be simulated using just the connectome.
When sugar-sensitive neurons in a fruit fly’s mouth (green) detect something sweet, they send signals to the brain, activating other neurons (light brown). Some of those neurons stimulate motor neurons to extend the proboscis and suck up the sugar. Image provided: Philip Shiu, PhD
“It’s been unclear how much the connectome would actually allow us to predict neural activity,” said Shiu, who currently works at , a startup that develops AI approaches to modeling connectomes. “Now, we and others have found that the connectome really does critically allow us to predict and understand how the brain works.”
While simple in concept, it was not at all clear whether such an approach could succeed. To test it out, Shiu built a connectome-based computational model of the brain, which models brain-wide neural firing. He then worked with Sterne and others to test out the model using well-described circuits for feeding and grooming behavior. They found that the model’s predictions matched up with what was already known about these circuits.
The researchers then took the model one step further. Since the sensory inputs and neuronal outputs that trigger feeding behavior are already well understood, they wondered whether the model could be used to predict how information about different tastes flows through the brain.
It surprised the scientists to learn that the model predicted overlap in circuitry for different tastes–such as sugar versus water. “We didn’t expect the overlap predicted by the model, but were able to experimentally confirm these predictions because the FlyWire Connectome and the model helped us pinpoint where to look in the Drosophila brain. These findings showed the model is also able to predict features of circuits that are non-intuitive, which can then be confirmed experimentally.” In other words, the model can point scientists in the best direction quickly and efficiently.
“Instead of doing a huge experimental screen involving many flies and extending over many months, you can use the model to predict the outcomes of the screen in silico. This approach can narrow down the possible candidates from thousands to 10 or so in a fraction of the time it would take to complete the experimental screen,” Sterne said. “It really can help you move faster and predict how certain circuit arrangements are functioning.”
Putting the Connectome to use
Using the connectome to understand more about simple behavior, researchers at Max Planck Florida Institute for Neuroscience and Sterne collaborated on a study focused on stopping. For example, when that unwelcomed visitor enters your kitchen and stops walking to nibble on a piece of ripe fruit or to groom (yes, the fruit fly keeps itself clean)–what does that look like in the brain?
“Before this research we did not have a great understanding of how the brain accomplishes stopping at a neural level,” Sterne, PhD, co-author of this paper out in that found stopping for food looks quite different at a cellular level in the Drosophila brain than stopping to groom. “This paper represents a big paradigm shift because, instead of focusing on individual neuronal cell types, we have been able to identify circuit mechanisms”, in other words, how groups of neurons wire together into circuits to carry out a specific function or behavior.
Researchers identified two circuit mechanisms that fruit flies use to stop walking: the walk-off mechanism, and the break mechanism.
- Walk-OFF Mechanism – this is used when the fruit fly stops walking for food. Researchers found that inhibitory neurons impede descending commands that are sending walking signals from the brain to the body. This essentially turns off walking signals. Researchers also found that disrupting this mechanism causes flies to overshoot food when wanting to feed.
- Brake Mechanism – this is used when the fruit fly stops walking to groom. Researchers found that instead of only inhibiting walking, this mechanism works in the spinal cord of the fly to arrest stopping movements using an excitatory neuron type that indirectly inhibits neurons that activate walking while activating neurons that lock the legs into a stable posture.
“Essentially what the brake mechanism is doing is increasing resistance at the leg joints to produce a very stable posture,” Sterne explained. “This is in the context of grooming behavior, where you need to be able to both stop and lift one leg off the ground without falling over to groom the body. So the brake mechanism uses excitatory neurons to turn off ongoing walking commands and arrest stepping movements.”
The Drosophila model system has long been invaluable model system to the field of neuroscience and has provided a broader understanding of human biology. These latest findings suggest a generic circuit mechanism that may also be used to trigger halting or stopping in humans. These insights could also inform the design of robots and inspire new computing architectures.
But Sterne cautions that there is still a long road until a fly can be booted up to fly inside a computer. “We’re not there yet because one thing this connectome lacks is information about how the motor neurons connect to physical features of the body like the muscles.”
The Flywire consortium research was supported by the National Institutes of Health BRAIN Initiative, Wellcome, Medical Research Council, Princeton University, and the National Science Foundation.
