Drosophila melanogaster, the common fruit fly, is in some ways a simple creature. But in others, it’s so complex that, like with any life form, we’re only getting to the surface. Researchers have taken an important step forward with D. melanogaster by creating the most accurate digital twins ever – at least in terms of how it moves and to a certain extent, why.
NeuroMechFly, as the researchers at EPFL call their new model, is a “morphologically realistic biomechanical model” based on careful scanning and close observation of actual flies. The result is a 3D model and motion system that, when prompted, does things like walk around or respond to some basic stimulus just like a real fly.
Just to be clear, this is not a complete cell-by-cell simulation, but we have seen some progress over the last few years with much smaller microorganisms. It doesn’t simulate hunger, sight, or any complex behavior – not even the way it flies, just the way it walks along surfaces and masters itself.
What’s hard about that, you ask? Well, one thing is to estimate this type of motion or behavior and create a lifelike little 3D flying machine. It’s another way to do so with a degree of precision in a simulated physics environment, which includes biologically precise exoskeletons, muscles, and neural networks similar to those of the fly that controls them. .
To create this very precise model, they started with a CT scan of a fly, to create a morphologically realistic 3D mesh. They then recorded a fly walking under very controlled circumstances and tracked its precise foot movements. EPFL . researchers then need to model exactly how those movements correspond to “physically simulated articulated body parts, such as head, legs, wings, abdominal segments, proboscis, antennae, cords” , the latter is a kind of motion-sensing organ that helps in flight.
Image credits: Pavan Ramdya (EPFL)
They show that these movements work by bringing the precise movements of the observed fly into the simulated environment and replaying them with the simulated fly – real movements are precisely mapped onto the tissue. Figure. They then demonstrated that they could create new gait and movements based on these, so that the fly would run faster or more consistently than they had observed.
Image credits: Pavan Ramdya (EPFL)
Not that they’re improving in essence, to be exact; they just show that the fly motion simulation is extended to other, more extreme examples. Their model is even strong against virtual bullets… to a certain extent, as you can see in the animation above.
“These case studies have built our confidence in the model. But we were most interested in the moment when the simulation failed to reproduce the animal’s behavior, showing how to improve the model,” said Pavan Ramdya of EPFL, who led the team that built the simulation (and other models). related to D. melanogaster) said. Seeing where their simulation breaks shows where there’s work to be done.
“NeuroMechFly could advance our understanding of how behaviors emerge from the interaction between complex neuromechanical systems and their physical surroundings,” reads the paper’s abstract. published last week in the journal Nature Method. By better understanding how and why a fly moves the way it does, we can also better understand the systems that underpin it, generating insights into different areas. another area (fruit flies are one of the most used laboratory animals). And of course if we want to create an artificial fly for some reason, we will definitely want to know how it works first.
While NeuroMechFly is in some ways a giant step forward in the field of digital life simulation, it is still (as its creators will be the first to admit) extremely limited, focus only on specific physical processes and not on many other aspects of the tiny body. and the mind turns Drosophila into Drosophila. You can view the code and can contribute at GitHub or Ma Duong.
