Home

Sun-Nanometer Precision using Bayesian Grouping of Localizations (BaGoL): Single molecule localization microscopy super-resolution methods such as DNA-PAINT and (d)STORM generate multiple observed localizations over the time course of data acquisition from each dye or binding site that are nor a priori assigned to those specific dyes or binding sites. BaGoL implements a Bayesian method of grouping and combining localizations from multiple blinking/binding events that can improve localization precision to better than one naometer. BaGoL allows inclusion of prior knowledge such as distribution of the number of localizations per emitter and the localization precisions.

The algorithm is comprised of several steps depicted in the figure. First, the list of localizations are split into smaller subsets. Second, the outliers are recognized as localizations with less than a certain number of neighbors within a certain distance. Third, localizations within each subset are further split into preclusters using hierarchical clustering algorithm. Fourth, each precluster is processed using RJMCMC. Fifth, the chain from all the preclusters are combined to produce posterior and MAPN images.

We tested several other common algorithm for the porpuse of grouping and combining of the localizations and BaGoL did better than all of them. This method can be used for about a factor of two precision improvement on a typical dSTORM data set and facilitate further quantitative analysis. When using DNA-PAINT, the method can achieve better than one nanometer precision. We concieve numerous biological applications of the algorithm, such as inspection of protein-protein interactions, etc.