Towards Unbiased Fluorophore Counting in Superresolution Fluorescence Microscopy.

With the advent of fluorescence superresolution microscopy, nano-sized structures can be imaged with a previously unprecedented accuracy. Therefore, it is rapidly gaining importance as an analytical tool in the life sciences and beyond. However, the images obtained so far lack an absolute scale in terms of fluorophore numbers. Here, we use, for the first time, a detailed statistical model of the temporal imaging process which relies on a hidden Markov model operating on two timescales. This allows us to extract this information from the raw data without additional calibration measurements. We show this on the basis of added data from experiments on single Alexa 647 molecules as well as GSDIM/dSTORM measurements on DNA origami structures with a known number of labeling positions.