Optical microscopy currently stands as the most pivotal tool in biological sciences. Thanks to the ingenuity and creativity of human researchers, super-resolution (SR) methods have emerged, breaking through the classical diffraction limit of light—around 250 nm—and allowing us to explore the intricate organization of the smallest functional units of cellular life.
Traditionally, discovering new microscopy techniques has depended heavily on human experience, intuition, and creativity—a challenging task considering the vast array of potential experimental optical configurations. For instance, with an optical setup composed of just 10 elements selected from 5 different components, like mirrors, lenses, or beam splitters, there could be over 100 million unique configurations.
The complexity of this scenario indicates that many effective techniques may still be undiscovered, suggesting that human intuition alone may not suffice. This presents a significant opportunity for AI-driven exploration techniques, which could navigate this complex space quickly and without bias.
“Experiments are our windows to the Universe, into the large and small scales. Given the sheer enormously large number of possible experimental configurations, it’s questionable whether human researchers have already discovered all exceptional setups. This is precisely where artificial intelligence can help,” explains Mario Krenn, head of the “Artificial Scientist Lab” at MPL.
To address this challenge, scientists from the “Artificial Scientist Lab” joined forces with Leonhard Möckl, a domain expert in super-resolution microscopy and head of the “Physical Glycoscience” research group at MPL. Together, they developed XLuminA, an efficient open-source framework designed with the ultimate goal of discovering new optical design principles.
The researchers utilize its features, particularly in the realm of SR microscopy. XLuminA functions as an AI-based optics simulator that can autonomously assess the full range of possible optical configurations. What distinguishes XLuminA is its remarkable efficiency: it utilizes advanced computational techniques to evaluate potential designs at a speed 10,000 times greater than conventional computational methods.
“XLuminA is the first step towards bringing AI-assisted discovery and super-resolution microscopy together. Super-resolution microscopy has enabled revolutionary insights into fundamental processes in cell biology over the past decades — and with XLuminA, I’m convinced that this story of success will be accelerated, bringing us new designs with unprecedented capabilities,” adds Leonhard Möckl, head of the “Physical Glycoscience” group at MPL.
The first author of the work, Carla Rodríguez, along with her team, validated their approach by showing that XLuminA could independently rediscover three foundational microscopy techniques. They started with simple optical configurations, successfully rediscovering a system designed for image magnification.
Moving on to more intricate challenges, the researchers adeptly rediscovered the prestigious STED (stimulated emission depletion) microscopy, along with a method for achieving super-resolution (SR) utilizing optical vortices. The pinnacle of their work showcases XLuminA’s remarkable ability for true discovery.
By challenging the framework to find the optimal SR design with the available optical elements, the team witnessed an extraordinary outcome: XLuminA independently synthesized a way to combine the foundational principles of both STED microscopy and the optical vortex technique into a single, novel experimental blueprint. This new design not only stands out as a previously unreported innovation but also surpasses the performance of each individual SR technique.
“When I saw the first optical designs that XLuminA had discovered, I knew we had successfully turned an exciting idea into a reality. XLuminA opens the path for exploring completely new territories in microscopy, achieving unprecedented speed in automated optical design. I am incredibly proud of our work, especially when thinking about how XLuminA could help in advancing our understanding of the world. The future of automated scientific discovery in optics is truly exciting!” says Carla Rodríguez, the study’s lead author and main developer of XLuminA.
The framework’s modular design facilitates easy adaptation to various microscopy and imaging methods. In the future, the team intends to incorporate nonlinear interactions, light scattering, and temporal data, allowing for the simulation of systems like iSCAT (interferometric scattering microscopy), structured illumination, and localization microscopy, among others.
This framework can be utilized by other research teams and tailored to meet their specific requirements, providing significant benefits for interdisciplinary research partnerships.
Journal reference:
- Carla Rodríguez, Sören Arlt, Leonhard Möckl, Mario Krenn. Automated discovery of experimental designs in super-resolution microscopy with XLuminA. Nature Communications, 2024; DOI: 10.1038/s41467-024-54696-y