A coordinate-based co-localization index to quantify and visualize spatial associations in single-molecule localization microscopy

Visualizing the subcellular distribution of proteins and determining whether specific proteins co-localize is one of the main strategies in determining the organization and potential interactions of protein complexes in biological samples. The development of super-resolution microscopy techniques such as single-molecule localization microscopy (SMLM) has tremendously increased the ability to resolve protein distribution at nanometer resolution. As super-resolution imaging techniques are becoming instrumental in revealing novel biological insights, new quantitative approaches that exploit the unique nature of SMLM datasets are required. Here, we present a new, local density-based algorithm to quantify co-localization in dual-color SMLM datasets. We show that this method is broadly applicable and only requires molecular coordinates and their localization precision as inputs. Using simulated point patterns, we show that this method robustly measures the co-localization in dual-color SMLM datasets, independent of localization density, but with high sensitivity towards local enrichments. We further validated our method using SMLM imaging of the microtubule network in epithelial cells and used it to study the spatial association between proteins at neuronal synapses. Together, we present a simple and easy-to-use, but powerful method to analyze the spatial association of molecules in dual-color SMLM datasets.