Scientists Create First Digital Twin of a Fruit Fly, Opening New Frontiers in Robotics and Neuroscience
Researchers at the Swiss Federal Institute of Technology Lausanne (EPFL) have achieved a milestone in computational neuroscience and bio-inspired engineering: the creation of NeuroMechFly, the first fully functional digital twin of the common fruit fly, Drosophila melanogaster. The breakthrough, published in Nature Methods, marks a significant leap toward reverse-engineering the neural and mechanical control of animal behavior—and could pave the way for a new generation of agile, insect-like robots.
“We used two key data sources to build NeuroMechFly,” explained Pavan Ramdya, a professor at EPFL’s School of Life Sciences and the project’s lead researcher. “First, we performed a CT scan of an actual fly to construct a morphologically realistic biomechanical model. Second, we tracked the precise movements of the fly’s limbs using pose-estimation software we developed over the past few years.” The result is a digital model that doesn’t just mimic the fruit fly’s anatomy—it replicates its movements with lifelike precision.
The fruit fly has long been a cornerstone of biological research, thanks to its relatively simple nervous system and well-documented genetics. But NeuroMechFly goes beyond traditional biological study. By combining high-resolution imaging with advanced motion-capture technology, the EPFL team has created a tool that allows scientists to simulate and analyze the fly’s behavior in ways that were previously impossible. The implications stretch far beyond the lab: this digital twin could aid engineers design robots that move with the same efficiency and adaptability as insects, a goal that has eluded roboticists for decades.
The project didn’t happen overnight. Ramdya and his team have spent years refining their approach, collaborating with EPFL’s Biorobotics Laboratory under Professor Auke Ijspeert. Their work builds on a growing trend in computational neuroscience—using digital models to bridge the gap between biological systems and artificial intelligence. NeuroMechFly isn’t just a static simulation; it’s a dynamic platform that can be used to test hypotheses about how the fly’s brain controls its body, offering insights that could one day be applied to everything from prosthetics to autonomous drones.
“NeuroMechFly represents a major step toward reverse-engineering the neuro-mechanical control of animal behavior,” Ramdya said. “It also opens doors for developing bio-inspired robots that can navigate complex environments with the same agility as insects.”
While NeuroMechFly is the first of its kind, it’s not the only effort to digitize the fruit fly’s biology. Projects like Neurokernel, an open-source initiative aimed at emulating the entire fruit fly brain on GPUs, have laid the groundwork for large-scale neural simulations. But NeuroMechFly stands out for its focus on the intersection of neural control and physical movement—a critical piece of the puzzle for creating robots that can suppose and move like living organisms.
The potential applications are vast. In robotics, digital twins like NeuroMechFly could lead to machines that can crawl, climb, or even fly with the same energy efficiency and adaptability as insects. In neuroscience, the model offers a new way to study how the brain processes sensory information and translates it into movement—a question that remains one of the field’s biggest challenges. And in medicine, insights from the fly’s neural circuitry could inform the development of brain-machine interfaces or advanced prosthetics.
For now, NeuroMechFly is a research tool, but its creation signals a broader shift in how scientists approach the study of biology. By combining digital modeling with real-world data, researchers are no longer limited to observing life—they can simulate it, tweak it, and even reimagine it. As computational power continues to grow, digital twins of more complex organisms may not be far behind, bringing us closer to a future where the line between biology and technology blurs entirely.
The team’s findings were published on April 28, 2026, in Nature Methods, cementing NeuroMechFly’s place as a landmark achievement in both neuroscience and robotics. For engineers and scientists alike, the tiny fruit fly has just grow a giant leap forward.
