Most materials do what their composition tells them to do. Weizmann Institute of Science researchers have now built materials that follow a genetic program instead. In two studies announced on November 9, 2026, the lab of Dr. Alexandra Tayar showed that DNA can switch a moving, living-like material on, pace its motion, and even synchronize it across millimeters.
The work, published in Nature Materials and Nature Physics, targets a field called active matter, where tiny molecular motors burn chemical fuel and turn it into movement. The Weizmann team says it lays groundwork for genetically programmed materials, from soft robots to targeted therapies. These are early laboratory results, not products.
What active matter is, and why control has been hard
Active matter covers materials whose parts consume energy and move on their own, much like the swimming of bacteria or the flow of molecules inside a cell. According to the Weizmann release, scientists have studied how such systems organize for only about two decades, and have mostly steered them by changing the ingredients, such as the type and amount of microtubules and motor proteins.
That leaves a gap. A living cell senses its own state and adjusts. A lab-built active material, until now, mostly could not. Tayar set out to combine her two specialties: motor-driven active matter and artificial systems that express genes outside cells. The question was whether a material could be given a genetic program that lets it make the very motors that drive it.
Study one: a material that runs on a genetic code
In the Nature Materials study, led by doctoral student Rochelle Silverman and staff scientist Dr. Erez Zerbib, with Prof. Vincent Noireaux of the University of Minnesota, the team embedded DNA and a gene expression system inside a network of microtubules. The material could then manufacture its own motor proteins from the instructions in the DNA.
When the researchers let a motor gene express continuously, motor concentration rose steadily. Once motor activity passed the threshold needed to overcome the network’s elastic forces, the material “came to life.” Microtubules began to move and form protrusions that grew larger, and eventually the network broke into bundles that swirled in small vortices for hours.
Varying the amount of the starting gene changed when the material started moving, how fast it moved, and how quickly it shifted between mechanical states. To control where protrusions formed, the team used 3D printers to create microscopic channels and ridges around the material, adding a geometric layer of control.
The group then wrote small programs. A gene for a bridging protein, which Silverman likens to molecular Velcro, made the material move briefly and then stop as friction built up between tubes. Adding another protein that shuts down motor production after a delay produced bursts of movement at regular intervals, with proteins rapidly degraded to make room for each new pulse. A simple theoretical model, developed with Dr. Hillel Aharoni, explained how gene expression rates translate into structural changes.
Study two: DNA as a molecular spring
The second study, published in Nature Physics and led by doctoral students Maya Levanon and Noa Goldberg, tackled a different puzzle: how countless microscopic interactions produce large-scale, coordinated motion. The idea came from Tayar’s earlier work on DNA-based force sensors, where she saw DNA strands stretch and relax at about the same rate the motors drove the material.
The team added long DNA strands to an active material. The vortex flows tangled the DNA into a three-dimensional network, and that network produced rhythmic oscillations. The denser and more entangled the network, the larger the oscillations. According to the researchers, the rhythm comes from the network’s delayed response: flow in one direction, say clockwise, stretches the network for about 10 minutes until it reaches its limit, and then it contracts and sweeps the material the other way.
Two further findings stand out. The DNA has to be long enough to bridge neighboring vortices for stable, synchronized motion to appear. And the millimeter-scale oscillation amplitude depended only on the imposed boundaries, growing in direct proportion to container size, while speed was set by motor concentration. The release adds that a minimum motor concentration is needed before the DNA entangles into a network. Theory partners included Prof. Eran Bouchbinder at Weizmann and Dr. Ram Adar and Dvir Cohen at the Technion.
What it means
Together, the studies give materials engineers and drug developers a toolkit of genetic and mechanical levers for designing how an active material behaves. Tayar says she hopes combining the tools will eventually produce soft robots that change shape at the right time, in the right place and in the right way, and therapies that reach target cells more efficiently.
There is also a biological hint. Tayar notes that bacteria, drugs and sperm cells move through polymer-rich mucus in the body. Intuition says long molecular chains should make swimming harder, but the findings suggest such chains may actually help synchronize motion and make it more efficient.
For innovation watchers, the pattern is worth noting: biology’s own control variables, such as the rate of gene expression or a genetic feedback loop, are being treated as physical design parameters for engineered matter. The release gives one telling number: fully stretching a single DNA molecule takes 1 to 65 piconewtons, a range that closely matches the force of the kinesin motor protein used in these materials.
Caveats
These are fundamental-science results from a laboratory, working with microtubules and motor proteins in controlled settings. The release describes applications such as soft robots and targeted therapies as possibilities the work could lead toward, and no product, clinical test, or timeline is announced. Nothing here is medical guidance.
Conclusion
By letting DNA decide when a material wakes up, how it pulses, and how its motion synchronizes, the Weizmann team has moved active matter a step closer to programmable design. It is a reminder that the next generation of smart materials may be written, quite literally, in code.