It started with M.C. Escher’s wood engraving Depth. In the early 1980s, Nadrian Seeman was looking for a solution to a crystallography problem. The symmetry and the spatial arrangements of the school of fish in Escher’s hypnotic panel got him thinking about DNA self-assembly, and the field of DNA nanotechnology was born1. Since then, the simple DNA topologies designed by Seeman and colleagues have evolved into complex DNA origami structures that can be programmed to adopt specific shapes and dynamic DNA machines that are capable of molecular sensing2.

DNA nanotechnology has so far provided significant fundamental insight into the capabilities of bottom-up molecular assembly, allowing exquisite control over the organization of material and molecules at the nanoscale. But the translation of DNA nanotechnology from an academic concept to a practical tool is still in its infancy. Now a series of papers showcasing micrometre-sized two-dimensional DNA arrays and gigadalton three-dimensional origami structures might move the field a little bit closer to this goal, proving that bigger and more complex architectures can be built with high yields and low error rates3,4,5. Notably, a smart one-pot biotechnological approach published alongside these papers also offers a pragmatic route to cut DNA origami production costs6. In this strategy, bacteriophages are programmed to amplify a precisely engineered, self-cleaving DNA template that generates both the scaffold and the staple sequences needed for ‘in-phage’ self-assembly. The authors of the study predict that the method, scaled up to a typical biotech pilot, will yield kilograms of origami at 0.18 euros per milligram — two to three orders of magnitude lower than the current costs.

Bigger structures are unlikely to be ideal for every practical implementation: with certain in vivo applications, for example, a minimalist approach could allow architectures to be designed that retain complex functions while being simple enough to show low toxicity in the body. But in other cases, such as for energy and photonic applications, large structures are potentially desirable. Moreover, exploring self-assembly at different length scales, from molecular complexes to virus-like vesicles and cell organelles, could unlock new possibilities in the emerging fields of artificial cells and molecular robotics.