The Impact of Image Quality on Reliable and Specific Colocalization Analysis

Abstract Colocalization analysis is a well-established method to quantify the extent to which fluorescent signal of two different objects or molecule populations are related. Irrespective of the chosen colocalization approach (correlation, co-occurrence, or object-based), the results are fiercely affected by factors such as image background, noise, crosstalk, chromatic aberration, and the diffraction barrier of the optical microscope. Diffraction leads to a blurry appearance of the captured image as each point emitter is represented by the point spread function (PSF) in the image. The PSF structure is also highly dependent on the wavelength of the fluorophore and the type of microscope used for imaging. For example, confocal imaging would show two-point emitting fluorophores as colocalized within its 200 nm lateral resolution limit due to its relatively large PSF, whereas array detector, SMLM, or STED imaging may resolve these same objects as two distinct points. With its capacity to increase the resolution of both super-resolution and conventional microscopy data, deconvolution offers similarly more defined colocalization results by improving resolution, correcting out-of-focus signal, and reducing noise. In this article, we summarize the imaging conditions affecting colocalization analysis with a strong focus on image deconvolution and restoration. We present a range of tools in Huygens Software™ to correct for these imaging aberrations so that, when collectively applied, the fidelity of colocalization analysis is significantly enhanced.