Giovanni Volpe watches from across a lab bench in Gothenburg as a laser beam, guided by invisible algorithms, plucks a single DNA molecule from a microscopic crowd—no human hands, no pauses, no fatigue. This is SmartTrap, an AI-powered system developed by researchers at the University of Gothenburg and Chalmers University of Technology that’s redefining what’s possible in biophysical research. By teaching artificial intelligence to operate optical tweezers—tools so precise they can manipulate objects a thousand times thinner than a human hair—the team has unlocked a new era of high-speed, autonomous experimentation.
Optical tweezers, which earned Arthur Ashkin the 2018 Nobel Prize in Physics, have long been essential for studying life at the molecular level. They’ve helped scientists understand how DNA repairs itself, how cells move, and how diseases like malaria alter red blood cells. But until now, their use has been bottlenecked by human operation: researchers manually guiding each experiment, making decisions in real time, and enduring long, repetitive hours at the microscope. Throughput has been low, variability high, and progress slow.
SmartTrap changes that. Built on open-source software and powered by real-time deep learning, customized electronics, and precise fluid control, the AI platform runs experiments around the clock. It captures microscopic particles, positions them with nanometer precision in three dimensions, performs measurements, and loads new samples—all without human intervention. In rigorous testing, SmartTrap sorted and characterized hundreds of particles per hour and conducted 10 to 15 single-molecule DNA stretching experiments each hour, one of the most technically demanding tasks in biophysics. It also mapped nanoscale electrostatic forces between particles at varying salt concentrations and measured the mechanical stiffness of red blood cells with remarkable consistency.
"We found that the AI performed as well as or better than a skilled human operator in every case," says Volpe. Unlike humans, the system doesn’t tire, lose focus, or need breaks. It simply works—accurately, tirelessly, and at speeds up to 100 times faster than traditional methods. The implications are profound: labs could accelerate drug discovery, deepen understanding of genetic diseases, and scale up nanoscale research in ways previously unimaginable.
Published in Nature Methods (2026, DOI: 10.1038/s41592-026-03129-3), SmartTrap is designed as a shared platform, inviting collaboration across academia and industry. As AI-driven microscopy matures, Volpe believes systems like SmartTrap will transform laboratories the way automation revolutionized manufacturing—quietly, efficiently, and irreversibly. The future of discovery isn’t just smarter. It’s already working the night shift.