Image processing and analysis for single-molecule localization microscopy: Computation for nanoscale imaging

Fluorescence microscopy is currently the most important tool for visualizing biological structures at the subcellular scale. The combination of fluorescence, which enables a high imaging contrast, and the possibility to apply molecular labeling, which allows for a high imaging specificity, makes it a powerful imaging modality. The use of fluorescence microscopy has risen tremendously, in particular since the introduction of the green fluorescent protein (GFP) in the mid-1990s and the possibility to genetically engineer cells to express these proteins. Figure 1 shows the basic layout of a fluorescence microscope. Excitation light of a certain wavelength is reflected via a dichroic beamsplitter and projected onto the specimen via the objective lens of the microscope. The light is absorbed by the fluorescent labels and re-emitted, slightly Stokes-shifted by ∼10-100 nm, at a larger wavelength, typically a few nanoseconds later. The emission light is captured by the objective lens and directed toward the camera via the dichroic beamsplitter.