Drosophila melanogaster, the common fruit fly, is in some ways a simple creature. But in others it is so complex that, as with any form of life, we are only scratching the surface of its understanding. But researchers have taken a big step forward with Drosophila, creating the most accurate yet digital twin — at least of how it moves and, to a certain extent, why.
NeuroMechFly, as EPFL researchers call their new model., is a “morphologically realistic biomechanical model” based on careful scanning and close observation of real flies. The result is a 3D model and movement system that, when asked to do something like walk or respond to certain basic stimuli, is very much like a real fly.
To be clear, this is not a full cell-by-cell simulation where we have seen some progress over the past few years with much smaller microorganisms. It doesn’t mimic hunger, vision, or any complex behavior – not even the way it flies, just the way it walks on the surface and grooms itself.
What’s so hard about this, you ask? Well, it’s one thing to approximate this type of movement or behavior and make a small 3D fly that moves more or less like a real one. It’s another thing to do it to the exact degree in a fully physically simulated environment, including a biologically accurate exoskeleton, muscles, and a fly-like neural network that controls them.
To make this highly accurate model, they started with a CT scan of a fly to create a morphologically realistic 3D mesh. They then recorded the fly walking under very carefully controlled conditions and tracked the movements of its legs very accurately. They then had to accurately model how these movements correspond to physically modeled “articulated body parts such as the head, legs, wings, abdominal segments, proboscis, antennae, halteres”, the latter of which is a kind of motion-sensing organ that aids in flight time.
They showed that this works by taking the exact movements of the observed fly into the simulation environment and replicating them with the simulated fly—the real movements are correctly displayed on the model. They then showed that they could create new gaits and movements based on them, allowing the fly to run faster or more stable than what they had observed.
It’s not that they’re improving nature, they’re just showing that the fly’s motion simulation extends to other, more extreme examples. Their model was even resistant to virtual projectiles… to a certain extent, as you can see in the animation above.
These case studies strengthened our confidence in the model. But we are most interested in when the simulation fails to replicate animal behavior, pointing out ways to improve the model,” said Pawan Ramdya of EPFL, who leads the team that created the simulation (and others). Drosophilarelated models). Seeing where their simulation fails reveals where more work needs to be done.
“NeuroMechFly can improve our understanding of how behavior emerges from the interaction between complex neuromechanical systems and their physical environment,” the paper’s abstract says. published last week in Nature Methods. By better understanding how and why the fly moves the way it does, we can better understand the underlying systems as well as gain insights in other areas (Drosophila is among the most commonly used experimental animals). And of course, if we ever wanted to create an artificial fly for some reason, we would definitely like to know how it works first.
Although NeuroMechFly is in some ways a huge advance in digital life simulation, it is still (as its creators are the first to admit) incredibly limited, focusing solely on specific physical processes rather than many other aspects of the tiny body. and remember what to do Drosophila a Drosophila. You can check the code and maybe contribute to Github or ocean code.
Credit: techcrunch.com /